system

Abstract

A system is provided.SOLUTION: A system comprising: means for constructing a distributed network; means for supplying computing resources of respective terminals; means for distributing an image generation task; means for processing the distributed image generation task and transmitting a generated image to a server; means for analyzing and sorting the generated image by AI; means for generating a high-precision image by using a GAN based on the sorted image; and means for calculating a reward and paying the reward to the terminals contributing to the generation.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional image generation AI requires a lot of trial and error and fine-tuning for users to obtain images that suit their purposes, which requires a lot of computational resources and time. Furthermore, centralized systems make it difficult to properly distribute rewards to the devices (users) that provide the computational resources. Furthermore, current image generation technologies sometimes lack accuracy and realism, making it difficult to meet user expectations. [Means for solving the problem]

[0005] The present invention builds a distributed network and provides a means for each terminal to supply computational resources. Using this network, a server distributes image generation tasks to each terminal, and the terminals send the generated images to the server. The server analyzes the received images using AI and selects them based on evaluation criteria such as texture and color. Then, a highly accurate image is generated using a GAN based on the selected images. The server also has a means for checking the distributed network log, calculating rewards based on the provision of computational resources, and paying rewards in electronic money to terminals that contributed to the generation. In this way, efficient and highly accurate image generation and fair reward distribution can be achieved.

[0006] A "distributed network" is a network in which each terminal (node) shares information with each other and operates cooperatively, without relying on a centralized server.

[0007] A "terminal" is a computer that participates in the distributed network, provides computing resources, and processes image generation tasks.

[0008] "Computational resources" refers to the computing power of a device, such as the CPU and GPU, that is used to process image generation tasks.

[0009] The "server" is a central computer that manages the distributed network, distributes image generation tasks, and collects and analyzes the generated images.

[0010] An "image generation task" refers to a specific image generation task that each terminal processes in a distributed network, such as generating a specific pattern.

[0011] "Analysis" means using AI algorithms to evaluate the generated images and select them based on evaluation criteria such as texture and color.

[0012] "Sorting" means selecting the best images that meet certain criteria from among those evaluated by an AI algorithm.

[0013] "GAN" stands for Generative Adversarial Network, a technology that generates highly accurate images by having two neural networks, a discriminator and a generator, compete to generate images.

[0014] "Reward" refers to the compensation paid to a terminal (user) that provides computing resources and contributes to image generation, and is provided in the form of electronic money or the like.

[0015] "Electronic money" means currency issued in digital form and used for reward payments on a decentralized network. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] As an embodiment of the present invention, a system for generating highly accurate images using a distributed network and paying rewards to terminals that contribute to the image generation will be described below. The system operates mainly in cooperation between a server, terminals, and users.

[0038] Server Features

[0039] 1. Network Construction

[0040] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[0041] 2. Task Distribution

[0042] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[0043] 3. Image collection and analysis

[0044] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[0045] 4. GAN Processing

[0046] The server uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs are made up of two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[0047] 5. Reward Allocation

[0048] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0049] Device Features

[0050] 1. Registering with the Network

[0051] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0052] 2. Receiving and processing tasks

[0053] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[0054] 3. Sending images

[0055] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[0056] User Roles

[0057] 1. Request for image generation

[0058] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[0059] 2. Receiving Rewards

[0060] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0061] Specific examples

[0062] Image generation request example

[0063] 1. Receiving a request

[0064] The server receives a request from a user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[0065] 2. Task Distribution and Processing

[0066] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[0067] 3. Image selection and GAN processing

[0068] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0069] 4. Reward Allocation

[0070] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0071] The above is a detailed description of the embodiment of the present invention. This system realizes efficient and highly accurate image generation and appropriate reward distribution using a distributed network.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The server creates a distributed network and configures it so that nodes (devices) can join, which includes setting up public IP addresses so that devices can connect to the network.

[0075] Step 2:

[0076] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[0077] Step 3:

[0078] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[0079] Step 4:

[0080] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[0081] Step 5:

[0082] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[0083] Step 6:

[0084] The devices process the received tasks and generate images. Device A generates an image with a sandy texture, while device B generates an image with a mountain silhouette.

[0085] Step 7:

[0086] The device sends the generated images to the server, and metadata such as computational resource usage is included with each image.

[0087] Step 8:

[0088] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[0089] Step 9:

[0090] The server then uses GAN to generate a highly accurate final image based on the selected images. GAN uses two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[0091] Step 10:

[0092] The server checks the distributed network logs and records which devices provided how much computing resources.

[0093] Step 11:

[0094] The server calculates rewards based on the provision of computing resources. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[0095] Step 12:

[0096] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[0097] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and achieves appropriate reward distribution.

[0098] Example 1

[0099] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0100] This invention relates to a system that efficiently realizes high-precision image generation and fairly distributes rewards to terminals that provide computational resources that contribute to image generation. Conventional systems mainly use a centralized approach, which has led to problems such as the concentration of computational resources and unfair distribution of rewards. Furthermore, high-precision image generation requires a large amount of computational resources, which calls for a distributed computing environment.

[0101] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0102] In this invention, the server includes a means for constructing a distributed network, a means for providing computational resources to each terminal, and a means for receiving and decomposing image generation requests. This enables efficient high-precision image generation and fair distribution of rewards to terminals that provide computational resources. Specifically, the server assigns image generation tasks to each terminal, evaluates the generated images, and generates a high-precision final image. The server also checks the distributed network log, calculates rewards based on the provision of computational resources, and pays them in electronic money, thereby achieving overall system efficiency and fairness.

[0103] A "distributed network" is a network in which terminals with multiple computing resources are connected on an equal basis and perform data processing and communication without central control.

[0104] "Computational resources" is a general term for hardware resources used for data processing and calculations, such as a terminal's CPU, GPU, and memory.

[0105] An "image generation request" is a request or prompt that a user sends to a server to generate a particular image.

[0106] An "image generation task" is a small unit of work required to realize an image generation request, and is distributed to each terminal.

[0107] "Generative network" refers to a generative adversarial network (GAN), an algorithm in which two neural networks, a discriminator and a generator, compete to generate data.

[0108] "Task allocation" refers to the operations and processes for appropriately distributing image generation tasks to each terminal.

[0109] "Reward" refers to the compensation paid to the terminal that contributed to the image generation, and is usually provided in the form of electronic money.

[0110] "Electronic money" means monetary value provided in digital form and used for online transactions and payment of rewards.

[0111] The present invention provides a system that uses a distributed network to achieve high-precision image generation and fairly distributes rewards to terminals that contribute to image generation. Specific embodiments for implementing this system are described below.

[0112] Hardware and Software Use

[0113] In this system, the server uses dedicated software to build and manage the distributed network. Specifically, distributed network software such as Apache Kafka and Ethereum is used. The server has computational resources such as CPU, GPU, and memory, and functions as a system to manage these. The terminals are computers or smart devices that provide the computational resources. The terminals use local resources (CPU / GPU) to process image generation tasks.

[0114] Processing flow

[0115] 1. Network Construction

[0116] The server builds a distributed network and sets up a public IP address to provide an environment in which nodes (terminals) can participate. The server also authenticates nodes and allows them to connect to the network safely.

[0117] 2. Receiving an image generation request

[0118] The user sends a request to the server to generate an image. For example, the user can send a prompt such as "I want you to generate a sunset ocean image."

[0119] 3. Task Decomposition and Allocation

[0120] The server receives an image generation request and divides it into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset." The server distributes these tasks according to the resource availability of each device.

[0121] 4. Task processing and image generation

[0122] The device processes tasks assigned by the server. For example, device A receives the task of "generating sky color" and uses its GPU to generate realistic sky colors. The generated partial images are stored in local storage.

[0123] 5. Image Collection and Evaluation

[0124] The server collects the partial images sent from each device, evaluates them using AI algorithms (such as ResNet or Inception), and selects the best images based on criteria such as texture and color.

[0125] 6. GAN processing and final image generation

[0126] The server generates a highly accurate final image based on the selected images using a Generative Adversarial Network (GAN), which consists of two neural networks, a discriminator and a generator, that learn by competing with each other.

[0127] 7. Reward Allocation

[0128] The server checks the distributed network log and collates which terminal provided how much computing resources. Based on this, it pays the user who provided the terminal a reward in the form of electronic money. The reward is transferred to the user's electronic wallet.

[0129] Specific examples

[0130] Prompt Sentence Examples

[0131] "I want to generate a sunset ocean"

[0132] In this way, our system can efficiently generate highly accurate images and distribute rewards fairly based on the computing resources provided by the terminals.

[0133] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0134] Processing Steps

[0135] Step 1: Building the network

[0136] 1. Specific actions

[0137] To build a distributed network, the server must install and configure dedicated software (e.g., Apache Kafka or Ethereum), set up a public IP address, and create an environment where connections can be made.

[0138] 2. Input

[0139] Necessary network configuration information, software configuration files

[0140] 3. Output

[0141] A fully distributed network, with devices ready to connect

[0142] 4. Detailed explanation

[0143] The server sets up a terminal authentication system to ensure that newly joining terminals can securely connect to the network. To maintain the health of the network, the server monitors the node status in real time.

[0144] Step 2: Receiving an image generation request

[0145] 1. Specific actions

[0146] The user sends a request to the server to generate an image. For example, the user inputs a prompt such as "I want to generate a sunset ocean image."

[0147] 2. Input

[0148] User prompt (e.g. "I want to generate a sunset ocean")

[0149] 3. Output

[0150] Image generation requests recorded in the server's internal database

[0151] 4. Detailed explanation

[0152] The server analyzes the prompt received from the user and records its contents in an internal database. Based on this information, task decomposition is performed in the next processing step.

[0153] Step 3: Break down and assign tasks

[0154] 1. Specific actions

[0155] The server divides the image generation request into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset," and assigns each task to a device.

[0156] 2. Input

[0157] Image generation request from user, device resource information

[0158] 3. Output

[0159] Individual image generation tasks distributed to devices

[0160] 4. Detailed explanation

[0161] The server allocates tasks efficiently, taking into account the resources of each device (CPU, GPU, memory usage), and notifies the device of the allocated tasks.

[0162] Step 4: Receiving and Processing Tasks

[0163] 1. Specific actions

[0164] The device receives tasks assigned by the server and adds them to a local queue. The device executes the tasks using the GPU or CPU to generate images.

[0165] 2. Input

[0166] Image generation tasks assigned by the server

[0167] 3. Output

[0168] Generated partial images, metadata (resources used, processing time, etc.)

[0169] 4. Detailed explanation

[0170] The device uses a local AI model (e.g., CNN or Transformer) based on the task content to generate the specified image elements, and the generated data is stored in local storage.

[0171] Step 5: Collect and evaluate images

[0172] 1. Specific actions

[0173] The server receives the partial images generated by each device and evaluates them using an AI algorithm (e.g., ResNet, Inception).

[0174] 2. Input

[0175] Partial images and metadata sent from the device

[0176] 3. Output

[0177] Evaluation results and selected optimal images

[0178] 4. Detailed explanation

[0179] The server analyzes the texture, color, sharpness, etc. of the collected images and selects the best images based on the evaluation criteria. This selection is used for the next GAN processing.

[0180] Step 6: GAN processing and final image generation

[0181] 1. Specific actions

[0182] The server uses the selected images as input data and generates a highly accurate final image using a Generative Adversarial Network (GAN).

[0183] 2. Input

[0184] Evaluated and selected optimal images

[0185] 3. Output

[0186] High-precision final image

[0187] 4. Detailed explanation

[0188] GAN involves two neural networks, a discriminator and a generator, competing to learn and ultimately generating highly accurate, realistic images, which are then stored in server storage.

[0189] Step 7: Reward Allocation

[0190] 1. Specific actions

[0191] The server checks the distributed network log and calculates the contribution of each device, and based on this, it pays electronic money to the users who provided their devices.

[0192] 2. Input

[0193] Distributed network log, contribution data of each device

[0194] 3. Output

[0195] Rewards sent to the user's electronic wallet

[0196] 4. Detailed explanation

[0197] The server analyzes the usage status of computing resources and processing results of each device, and calculates rewards based on that. The calculated rewards are transferred to the user's electronic wallet as electronic money, realizing fair reward distribution according to the degree of contribution.

[0198] (Application example 1)

[0199] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0200] The problem that the present invention aims to solve is to provide a system that can perform product quality inspections efficiently and with high accuracy in the manufacturing industry. Conventional product inspection methods have limitations in inspection efficiency and accuracy, and require a great deal of effort and cost. Furthermore, conventional methods often involve the subjective judgment of the inspector, making consistent quality control difficult.

[0201] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0202] In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and sending the generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, and means for using the generated images for product inspection and returning the inspection results to the robot. This enables highly accurate and efficient product quality inspection in the manufacturing industry.

[0203] A "distributed network" is a network in which multiple terminals provide computing resources in a distributed manner and work together to process a single task.

[0204] "Computational resources" refer to computer resources such as CPU and GPU performance, memory capacity, and storage capacity provided by a terminal.

[0205] An "image generation task" is a unit of work that includes specific calculations and processing to generate an image.

[0206] "AI" is an abbreviation for artificial intelligence, a technology that uses algorithms such as machine learning and neural networks to analyze data and behave intelligently.

[0207] "Selection" is the process of selecting the best images from the generated images based on evaluation criteria such as texture and color.

[0208] "GAN" is an abbreviation for Generative Adversarial Network, an algorithm in which two neural networks (a generator and a discriminator) compete with each other to learn and generate highly accurate data.

[0209] "Remuneration" is the compensation paid to a terminal that provides computing resources within the system, and is often in the form of electronic money.

[0210] "Product inspection" is the process of checking the quality and condition of manufactured products to ensure they meet standards.

[0211] A "robot" is a device that operates mechanically to perform designated tasks. It is often used in automated manufacturing processes.

[0212] MODE FOR CARRYING OUT THE INVENTION

[0213] As an embodiment of the present invention, a system for performing high-precision product inspection in a factory using a distributed network will be described. This system operates mainly in cooperation between a server, a terminal (robot), and a user.

[0214] Server Features

[0215] 1. Networking:

[0216] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[0217] 2. Task distribution:

[0218] The server receives product inspection requests from users and breaks them down into smaller tasks, such as inspecting specific parts or checking specific details. It then distributes these tasks to each terminal and manages them appropriately.

[0219] 3. Image Collection and Analysis:

[0220] The server collects product images acquired from each device, analyzes and evaluates the collected images using AI algorithms, and selects the most suitable images based on criteria such as the presence or absence of defects.

[0221] 4. GAN Processing:

[0222] The server uses GAN (Generic Annotation) to generate a highly accurate final image based on the selected images. GAN consists of two neural networks, a discriminator and a generator, that compete with each other to learn and provide more accurate inspection results.

[0223] 5. Reward Allocation:

[0224] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0225] 6. Return of test results:

[0226] The server compiles the test results and sends the final test results back to the robot.

[0227] Terminal (robot) functions

[0228] 1. Registering with the Network:

[0229] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0230] 2. Receiving and processing tasks:

[0231] The terminal receives and processes image generation and inspection tasks assigned to it by the server, performing specific tasks such as capturing images to inspect a specific part or evaluating the generated images to identify specific details.

[0232] 3. Sending images:

[0233] The device captures an image of the product and sends it to a server, along with metadata such as computing resource usage.

[0234] User Roles

[0235] 1. Product Inspection Request:

[0236] The user requests a specific product inspection from the server, specifying, for example, "Please inspect the surface of the part for scratches or dirt" or "Please check the product for color unevenness or shape abnormalities."

[0237] 2. Receiving rewards:

[0238] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0239] Hardware and software used

[0240] Hardware:

[0241] A robot installed on a factory production line (equipped with an image capture camera)

[0242] Modem / router for network connection

[0243] software:

[0244] Implemented server and robot control programs using Python.

[0245] Use the Requests library to process HTTP requests.

[0246] Use a local server or a cloud server (e.g. AWS, Google Cloud) on the server side.

[0247] Examples:

[0248] For example, one product inspection request might be "surface scratch inspection." The server splits this request into two tasks: "capturing an image of a specific part" and "analyzing the captured image using AI." The captured image is sent to the server and analyzed by an AI algorithm. The analysis results are used to determine whether there are any surface scratches using a deep learning model, and the results are sent back to the robot. The robot then decides on the next action to take on the production line.

[0249] Example prompt: "Inspect the product surface for scratches or stains."

[0250] This system enables factories to automate product inspection with high precision and efficiency, dramatically improving quality control.

[0251] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0252] Step 1: Register with the network

[0253] Server: The server sets a public IP address and builds a distributed network. It prepares the network environment so that devices can join. The input is the public IP address and network setting information, and the output is a network environment that devices can connect to.

[0254] Device: The device accesses the server to register on the network. As registration information, it sends its device ID and performance (CPU / GPU specs). The input is the device ID and performance information, and the output is a network participation token issued by the server.

[0255] Step 2: Receiving and processing the image generation task

[0256] Server: The server receives product inspection requests from users and divides them into image generation tasks. Tasks include inspecting specific parts and checking details. The input is the inspection request from the user, and the output is the image generation tasks assigned to each terminal.

[0257] Terminal: The terminal processes tasks received from the server. For example, it takes an image of a specific part of a product and sends the captured data to the server. The input is the image generation task from the server, and the output is the captured product image.

[0258] Step 3: Submit your images and add metadata

[0259] Terminal: When generating product images, metadata such as the usage of computing resources is added to the images when they are sent to the server. The input is the captured product image and metadata, and the output is a data package sent to the server.

[0260] Server: The server stores the received images and metadata and proceeds to the next analysis step. The input is the data package sent from the device, and the output is the stored image data and metadata.

[0261] Step 4: Analyze and select images

[0262] Server: The server uses an AI algorithm to analyze the received images. It selects the most suitable images based on criteria such as texture and color. The input is the stored image data and metadata, and the output is the selected image.

[0263] Step 5: GAN Processing

[0264] Server: The server generates a highly accurate final image using a generative anti-theft network (GAN) based on the filtered image. The input is the filtered image, and the output is the highly accurate final image.

[0265] Step 6: Generate and return test results

[0266] Server: Generates product inspection results based on the final image and sends the results back to the terminal. The input is the final image generated with high accuracy, and the output is the inspection result data.

[0267] Terminal: The terminal processes the inspection results received from the server and executes the next instructions on the production line. The input is the inspection result data from the server, and the output is the instructions for the production line.

[0268] Step 7: Compensation and Allocation

[0269] Server: The server checks the distributed network log and calculates rewards based on the provision of computing resources by each terminal. The input is the network log and computing resource information, and the output is the reward amount for each terminal.

[0270] User: The user who provides the terminal receives the reward in the form of electronic money. The input is the reward information from the server, and the output is a transfer to the user's electronic wallet.

[0271] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0272] As an embodiment of the present invention, we will explain a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. The system mainly operates in cooperation between a server, terminals, and users.

[0273] Server Features

[0274] 1. Network Construction

[0275] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[0276] 2. Task Distribution

[0277] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[0278] 3. Image collection and analysis

[0279] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[0280] 4. Emotional Engine Adjustment

[0281] The server uses an emotion engine to analyze the user's emotions in real time and adjusts image generation parameters and task distribution based on the analysis.

[0282] 5. GAN Processing

[0283] The server then uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs use two neural networks, a discriminator and a generator, to compete with each other to learn and generate more realistic images.

[0284] 6. Reward Allocation

[0285] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0286] Device Features

[0287] 1. Registering with the Network

[0288] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0289] 2. Receiving and processing tasks

[0290] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[0291] 3. Sending images

[0292] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[0293] User Roles

[0294] 1. Request for image generation

[0295] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[0296] 2. Providing emotions

[0297] Users use cameras and sensors to send facial and voice data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[0298] 3. Receiving rewards

[0299] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0300] Specific examples

[0301] Image generation request example

[0302] 1. Receiving a request

[0303] The server receives a request from the user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[0304] 2. Task Distribution and Processing

[0305] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[0306] 3. Image selection and GAN processing

[0307] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0308] 4. Emotional Engine Adjustment

[0309] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. Based on this data, the server adjusts the parameters required for the next image generation.

[0310] 5. Reward Allocation

[0311] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0312] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[0313] The processing flow will be explained below.

[0314] Step 1:

[0315] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[0316] Step 2:

[0317] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[0318] Step 3:

[0319] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[0320] Step 4:

[0321] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[0322] Step 5:

[0323] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[0324] Step 6:

[0325] Terminal A and terminal B process the received task and generate images. Terminal A generates an image with a sand texture, and terminal B generates an image with a mountain silhouette.

[0326] Step 7:

[0327] Once the generated images are complete, device A and device B send each generated image to the server, along with metadata such as the usage of computing resources.

[0328] Step 8:

[0329] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[0330] Step 9:

[0331] The server uses GAN to generate a highly accurate final image based on the selected images. In this process, two neural networks, a discriminator and a generator, compete with each other to learn and generate realistic images.

[0332] Step 10:

[0333] While the user evaluates the generated images, real-time emotional data is collected via a camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction.

[0334] Step 11:

[0335] The server adjusts the parameters required for the next image generation based on the data analyzed by the emotion engine. For example, if the user is not satisfied, it adjusts the color tone or details.

[0336] Step 12:

[0337] The server checks the distributed network logs and records which devices provided how much computing resources.

[0338] Step 13:

[0339] The server calculates rewards based on the amount of computing resources provided. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[0340] Step 14:

[0341] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[0342] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and has an adjustment function that reflects the user's emotional data in real time, ultimately achieving appropriate reward distribution.

[0343] Example 2

[0344] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0345] Conventional image generation systems lack the technology to rapidly generate highly accurate images or to generate images that reflect user emotions. They also lack the means to efficiently utilize distributed networks and distribute appropriate rewards based on computational resources. To address these issues, a system that integrates advanced image analysis and generation technology with the ability to reflect users' real-time emotions is needed.

[0346] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0347] In this invention, the server includes a means for constructing a distributed network, a means for supplying computing resources to each terminal, and a means for distributing image generation tasks. This enables highly accurate image generation and further enables real-time analysis of user emotions and their reflection in image generation. It also makes it possible to allocate appropriate rewards to each terminal based on the provision of computing resources.

[0348] A "distributed network" is a network structure in which multiple terminals work together to process data and communicate without relying on a centralized server.

[0349] "Computational resources" refers to the hardware and performance required for data processing and calculations, such as a computer system's CPU, GPU, memory, and storage.

[0350] An "image generation task" is a specific task or instruction for generating a new image using a particular algorithm or process.

[0351] "AI analysis and selection" means using artificial intelligence technology to analyze input data or information and select the best option based on specific criteria.

[0352] An "emotion engine" is a technology that analyzes emotions from a user's facial expressions and voice and adjusts the system's operation based on the results.

[0353] "GAN" stands for Generative Adversarial Network, a technology in which two neural networks, a generator and a discriminator, learn by competing with each other to generate highly accurate data (such as images).

[0354] "Electronic money" is digital or virtual currency that has value in electronic form and is traded over the Internet, rather than physical cash.

[0355] A "log" is data that represents a record of system actions or events, which can later be used for analysis or auditing.

[0356] This invention is a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. This system operates mainly through the cooperation of three elements: the server, the terminal, and the user. The roles and specific operations of each are described below.

[0357] Server Features

[0358] 1. Network Construction

[0359] The server creates a distributed network and sets up a public IP address for devices to join, allowing them to contact the server and register on the network.

[0360] 2. Task Distribution

[0361] The server receives an image generation request from a user, divides the request into small tasks, and distributes them to each terminal. At this time, it assigns the task to the most suitable terminal depending on the type and difficulty of the task.

[0362] 3. Image collection and analysis

[0363] The server collects the images generated by each device and uses AI algorithms to analyze and evaluate them, including texture and color, to select the best images.

[0364] 4. Emotional Engine Adjustment

[0365] The server uses an emotion engine to analyze the user's emotion data in real time and adjusts the image generation parameters based on the results, making it possible to generate images that are in line with the user's emotions.

[0366] 5. GAN Processing

[0367] The server uses a generative adversarial network (GAN) to generate a highly accurate final image based on the selected images. GANs are made up of two neural networks, a generator and a discriminator, that compete with each other to learn and generate more realistic images.

[0368] 6. Reward Allocation

[0369] The server checks the distributed network log and pays each terminal electronic money in return for the computing resources provided.

[0370] Device Features

[0371] 1. Registering with the Network

[0372] The device accesses the server and registers its computing resources (CPU / GPU performance, etc.) After registration, the device operates as part of the distributed network and is ready to receive tasks.

[0373] 2. Receiving and processing tasks

[0374] The device receives image generation tasks assigned by the server and performs the tasks, such as "generating the texture of sand" or "generating a night view of a city."

[0375] 3. Sending images

[0376] When the generated image is sent to the server, metadata such as the usage status of computing resources is also sent.

[0377] User Roles

[0378] 1. Request for image generation

[0379] A user requests the server to generate a specific image, specifically, by describing the request, such as a "mountain landscape" or a "city night view."

[0380] 2. Providing emotions

[0381] Users use cameras and sensors to send their emotional data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[0382] 3. Receiving rewards

[0383] Users who provide their devices receive rewards based on the tasks their devices complete, and the rewards are transferred to the users' electronic wallets as electronic money.

[0384] Specific examples

[0385] Image generation request example

[0386] 1. Receiving a request

[0387] The server receives a request from a user, such as "I want you to generate a desert landscape." The request is divided into smaller tasks, such as "generate the texture of sand" and "generate distant mountains."

[0388] 2. Task Distribution and Processing

[0389] The server assigns the task of "generating a sand texture" to terminal A and the task of "generating a silhouette of a distant mountain" to terminal B. Terminal A generates the sand texture, and terminal B generates the mountain silhouette, and they each send them to the server.

[0390] 3. Image selection and GAN processing

[0391] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0392] 4. Emotional Engine Adjustment

[0393] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. The server then adjusts the parameters required for the next image generation based on this data.

[0394] 5. Reward Allocation

[0395] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0396] Prompt Sentence Examples

[0397] "I want a desert landscape to be generated."

[0398] "I want to see the city's night view."

[0399] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[0400] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0401] Step 1:

[0402] Server: Network construction

[0403] The server sets up a distributed network by setting up a public IP address, which allows devices to join the network. The input requires the server's network configuration information, and the output is an initialized network. Specifically, the server edits the network configuration file, defines the public IP address and port number, and then listens on the network.

[0404] Step 2:

[0405] End user: Input of image generation request

[0406] A user requests a server to generate a specific image. The input is the user's request (e.g., "desert landscape"), and the output is the request being processed by the server. Specifically, the user enters the statement "I want a desert landscape generated" into a form in a web application and presses the submit button.

[0407] Step 3:

[0408] Server: Analyzes the request and distributes tasks

[0409] The server analyzes the received request and divides the task into smaller parts. The input is the user's request, and the output is the divided tasks. Specifically, the server divides the task into tasks such as "generate the texture of sand" and "generate the silhouette of a distant mountain," and distributes each task to the terminal.

[0410] Step 4:

[0411] Terminal: Receiving and processing tasks

[0412] The device receives tasks from the server and generates images. The input is the task content distributed by the server (e.g., "generate the texture of sand"), and the output is the generated image. Specifically, the device uses a deep learning model to generate the texture of sand and create its image data.

[0413] Step 5:

[0414] Device: Send image

[0415] The device sends the generated image to the server. The input is the generated image data, and the output is the image data sent to the server. Specifically, the device uploads the generated image data to the API endpoint specified by the server.

[0416] Step 6:

[0417] Server: Image collection and analysis

[0418] The server collects images received from the devices and analyzes them using an AI algorithm. The input is the image data sent from each device, and the output is the evaluated image data. Specifically, the server selects the most suitable images based on evaluation criteria such as the texture and color of each image.

[0419] Step 7:

[0420] Server: Coordination by emotion engine

[0421] The server analyzes the user's emotional data using an emotion engine and adjusts the image generation parameters. The input is the user's emotional data (for example, facial expression data and voice data), and the output is the adjusted parameters. Specifically, the server collects the user's emotional data, analyzes it using the emotion engine, and adjusts the parameters of the image generation algorithm based on the results.

[0422] Step 8:

[0423] Server: High-precision image generation using GAN processing

[0424] The server generates a high-precision image using a GAN model based on the selected images. The input is the evaluated image data, and the output is the final high-precision image. Specifically, the server inputs the selected image data into the generator, and has it compete with the discriminator to progress learning, ultimately generating a high-precision image.

[0425] Step 9:

[0426] Server: Reward Distribution

[0427] The server checks the distributed network log and calculates rewards based on the provision of computing resources. The input is the processing log data of each terminal, and the output is the reward amount for each terminal. Specifically, the server analyzes the processing log of each terminal and transfers the calculated electronic money as a reward to the electronic wallet of each terminal owner.

[0428] (Application example 2)

[0429] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0430] Conventional image generation systems require large amounts of computational resources to generate highly accurate images, and they have difficulty reflecting the specific emotional satisfaction desired by users in real time. Furthermore, there is a lack of a way to effectively utilize user feedback in the image generation process. As a result, images that do not fully meet the user's requirements are often generated, and satisfactory results cannot be achieved.

[0431] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and transmitting generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, means for collecting real-time emotional data from users, analyzing it, and reflecting it in the image generation process, and means for linking images generated via a terminal such as a smartphone with user emotional feedback. This makes it possible to generate high-precision images while also reflecting user emotional feedback in real time, thereby achieving image generation that greatly improves user satisfaction.

[0432] A "distributed network" is a network in which multiple devices are interconnected and can share and process data without centralized control.

[0433] "Computational resources" refers to the capabilities of hardware and software to perform computational processing, and specifically includes the performance of CPUs and GPUs.

[0434] "Image generation task" means a set of instructions for processing or computing required to generate a particular image.

[0435] A "server" is a central computer that processes and provides data in response to requests from multiple terminals within a network.

[0436] A "terminal" refers to a computer or device that is connected to a distributed network and provides computing resources.

[0437] "AI analysis and selection methods" refers to technology that uses artificial intelligence to evaluate collected images and select the most suitable ones.

[0438] "GAN" stands for generative adversarial network, a type of algorithm that generates realistic images by training two neural networks, a generator and a discriminator, to compete with each other.

[0439] "High-precision images" refer to detailed, high-quality images that closely resemble real-world scenes and objects.

[0440] "Electronic money" means currency provided in digital form, an asset that is exchanged electronically.

[0441] "Real-time emotional data" refers to information that collects the user's emotional state in real time and processes that data immediately.

[0442] "Emotional feedback" refers to the evaluation and response of a system based on emotional expressions and data provided by the user.

[0443] This invention is a system that uses a distributed network to generate highly accurate images, and also has the function of recognizing and reflecting user emotions in real time. The following is a specific embodiment of this invention.

[0444] Server Features

[0445] The server has the following main functions:

[0446] 1. Networking:

[0447] The server creates a distributed network and configures it so that each terminal can provide computing resources. It uses public IP addresses to provide an environment where terminals can easily connect.

[0448] 2. Task distribution:

[0449] The system divides image generation requests received from users into small tasks and distributes them to each device. For example, it gives detailed instructions such as "generate the texture of sand" or "generate the silhouette of a distant mountain" to each device.

[0450] 3. Image Collection and Analysis:

[0451] Images generated by each device are collected and analyzed and evaluated using AI algorithms, and the best images are selected based on evaluation criteria such as texture, color, and composition.

[0452] 4. Emotional engine regulation:

[0453] The system analyzes the user's emotional data in real time and reflects it at each stage of the generation process. This is achieved by analyzing the emotional information provided by the user through a camera or microphone.

[0454] 5. GAN Processing:

[0455] To generate the final high-precision image, a generative adversarial network (GAN) is used based on the selected images. GAN includes two neural networks: a generator and a discriminator.

[0456] 6. Reward Allocation:

[0457] The system checks the distributed network logs, calculates rewards based on the computing resources provided, and pays each device electronic money.

[0458] Device Features

[0459] The terminal has the following features:

[0460] 1. Registering with the Network:

[0461] They connect to a server, register with the decentralized network, provide computing resources, and receive tokens in return.

[0462] 2. Receiving and processing tasks:

[0463] It receives and processes image generation tasks from the server, and includes metadata such as the usage of computing resources when sending the generated images to the server.

[0464] User Roles

[0465] Users have the following roles:

[0466] 1. Image generation request:

[0467] Request the server to generate a specific image, specifying specific content such as a "mountain landscape" or a "city night view."

[0468] 2. Providing Emotion Data:

[0469] Cameras and sensors are used to provide facial and voice data in real time, which is analyzed by the emotion engine and reflected in image generation.

[0470] Specific examples

[0471] A user opens the app and requests that an image of a "summer beach sunset" be generated. The app collects the user's emotional data in real time via the camera and microphone. When the user expresses emotions such as "happy" or "moved" while looking at the image, that data is sent to the server and reflected in the image generation. Finally, an image that satisfies the user is generated and displayed on the smartphone.

[0472] An example prompt for a generative AI model is:

[0473] "Generate a summer seaside scene. I'd like the image to have a tranquil, relaxing atmosphere with the sound of the waves. Additionally, I'd like the sunset to be reflected on the water to enhance its beauty."

[0474] The above is an embodiment of the present invention. This system not only generates highly accurate images but also reflects real-time emotional feedback from the user, thereby improving user satisfaction.

[0475] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0476] Step 1:

[0477] Request Receipt

[0478] The server receives an image generation request from a user.

[0479] (Input) The user's request (e.g., "Summer sunset at the beach").

[0480] (Output) Initial settings for the image generation task.

[0481] (Specific operation) A user requests the generation of a specific image through the app, and the server receives the request, which may include a prompt for the generating AI model.

[0482] Step 2:

[0483] Task division

[0484] The server divides the image generation request into multiple subtasks.

[0485] (Input) The received image generation request.

[0486] (Output) A set of divided image generation tasks.

[0487] (Specific operation) The server breaks down the data into specific parts such as "generating the texture of sand" or "adjusting the color of the sunset" and converts each part into a processable format.

[0488] Step 3:

[0489] Task Distribution

[0490] The server distributes the divided tasks to the appropriate terminals.

[0491] (Input) A set of divided image generation tasks.

[0492] (Output) Tasks sent to each device.

[0493] (Specific operation) The server assigns tasks and sends instructions based on each terminal's available computing resources and past performance.

[0494] Step 4:

[0495] Receiving and processing tasks

[0496] The terminal processes the received task and sends the results to the server.

[0497] (Input) Image generation task sent from the server.

[0498] (Output) The generated subimage and metadata.

[0499] (Specific operation) Each device executes the received task, specifically generating the texture and color of a specific image, and sends this to the server along with metadata including the usage status of computing resources.

[0500] Step 5:

[0501] Image collection and analysis

[0502] The server collects images sent from each device and analyzes them using AI algorithms.

[0503] (Input) Partial images and metadata sent from each device.

[0504] (Output) The analyzed image data.

[0505] (Specific operation) The server evaluates the texture, color, composition, etc. of each image and selects the most suitable image using an AI algorithm.

[0506] Step 6:

[0507] Emotion data collection and analysis

[0508] The server collects and analyzes the user's real-time emotional data.

[0509] (Input) Emotion data from the user via camera and microphone.

[0510] (Output) Parsed emotion information.

[0511] (Specific operation) While the user is checking the generated image, facial expression and voice data is collected via the smartphone's camera and microphone, and the server analyzes this data using the emotion engine.

[0512] Step 7:

[0513] Reflecting and adjusting emotional data

[0514] The server adjusts image generation parameters based on the analyzed emotional information.

[0515] (Input) Analyzed emotion information.

[0516] (Output) Adjusted image generation parameters.

[0517] (Specific operation) The server adjusts the parameters of the generation process according to the user's positive or negative emotions. This adjustment ensures that the next task executed reflects the emotional data.

[0518] Step 8:

[0519] Final GAN ​​processing

[0520] The server generates the final image using GAN based on the selected images.

[0521] (Input) Selected partial images and adjusted parameters.

[0522] (Output) High-precision final image.

[0523] (Specific operation) Using a GAN (generative adversarial network) with a generator and discriminator, the final result is a highly accurate and realistic image.

[0524] Step 9:

[0525] Final image courtesy

[0526] The server provides the generated final image to the user.

[0527] (Input) High-precision final image.

[0528] (Output) The image that is displayed on the user device.

[0529] (Specific operation) The server sends the generated final image to the user's device, and the app displays it.

[0530] Step 10:

[0531] Reward Allocation

[0532] The server distributes rewards to the terminals that participate in the task processing.

[0533] (Input) Distributed network log and contribution data of each device.

[0534] (Output) Reward payment to each device.

[0535] (Specific operation) The server analyzes the logs of the distributed network, calculates rewards based on the degree of contribution, and pays electronic money to each terminal.

[0536] This concludes the explanation of the specific processing steps. This system enables highly accurate image generation and reflects the user's emotional feedback in real time.

[0537] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0538] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0539] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0540] [Second embodiment]

[0541] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0542] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0543] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0544] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0545] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0546] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0547] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0548] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0549] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0550] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0551] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0552] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0553] As an embodiment of the present invention, a system for generating highly accurate images using a distributed network and paying rewards to terminals that contribute to the image generation will be described below. The system operates mainly in cooperation between a server, terminals, and users.

[0554] Server Features

[0555] 1. Network Construction

[0556] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[0557] 2. Task Distribution

[0558] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[0559] 3. Image collection and analysis

[0560] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[0561] 4. GAN Processing

[0562] The server uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs are made up of two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[0563] 5. Reward Allocation

[0564] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0565] Device Features

[0566] 1. Registering with the Network

[0567] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0568] 2. Receiving and processing tasks

[0569] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[0570] 3. Sending images

[0571] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[0572] User Roles

[0573] 1. Request for image generation

[0574] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[0575] 2. Receiving Rewards

[0576] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0577] Specific examples

[0578] Image generation request example

[0579] 1. Receiving a request

[0580] The server receives a request from a user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[0581] 2. Task Distribution and Processing

[0582] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[0583] 3. Image selection and GAN processing

[0584] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0585] 4. Reward Allocation

[0586] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0587] The above is a detailed description of the embodiment of the present invention. This system realizes efficient and highly accurate image generation and appropriate reward distribution using a distributed network.

[0588] The processing flow will be explained below.

[0589] Step 1:

[0590] The server creates a distributed network and configures it so that nodes (devices) can join, which includes setting up public IP addresses so that devices can connect to the network.

[0591] Step 2:

[0592] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[0593] Step 3:

[0594] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[0595] Step 4:

[0596] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[0597] Step 5:

[0598] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[0599] Step 6:

[0600] The devices process the received tasks and generate images. Device A generates an image with a sandy texture, while device B generates an image with a mountain silhouette.

[0601] Step 7:

[0602] The device sends the generated images to the server, and metadata such as computational resource usage is included with each image.

[0603] Step 8:

[0604] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[0605] Step 9:

[0606] The server then uses GAN to generate a highly accurate final image based on the selected images. GAN uses two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[0607] Step 10:

[0608] The server checks the distributed network logs and records which devices provided how much computing resources.

[0609] Step 11:

[0610] The server calculates rewards based on the provision of computing resources. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[0611] Step 12:

[0612] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[0613] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and achieves appropriate reward distribution.

[0614] Example 1

[0615] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0616] This invention relates to a system that efficiently realizes high-precision image generation and fairly distributes rewards to terminals that provide computational resources that contribute to image generation. Conventional systems mainly use a centralized approach, which has led to problems such as the concentration of computational resources and unfair distribution of rewards. Furthermore, high-precision image generation requires a large amount of computational resources, which calls for a distributed computing environment.

[0617] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0618] In this invention, the server includes a means for constructing a distributed network, a means for providing computational resources to each terminal, and a means for receiving and decomposing image generation requests. This enables efficient high-precision image generation and fair distribution of rewards to terminals that provide computational resources. Specifically, the server assigns image generation tasks to each terminal, evaluates the generated images, and generates a high-precision final image. The server also checks the distributed network log, calculates rewards based on the provision of computational resources, and pays them in electronic money, thereby achieving overall system efficiency and fairness.

[0619] A "distributed network" is a network in which terminals with multiple computing resources are connected on an equal basis and perform data processing and communication without central control.

[0620] "Computational resources" is a general term for hardware resources used for data processing and calculations, such as a terminal's CPU, GPU, and memory.

[0621] An "image generation request" is a request or prompt that a user sends to a server to generate a particular image.

[0622] An "image generation task" is a small unit of work required to realize an image generation request, and is distributed to each terminal.

[0623] "Generative network" refers to a generative adversarial network (GAN), an algorithm in which two neural networks, a discriminator and a generator, compete to generate data.

[0624] "Task allocation" refers to the operations and processes for appropriately distributing image generation tasks to each terminal.

[0625] "Reward" refers to the compensation paid to the terminal that contributed to the image generation, and is usually provided in the form of electronic money.

[0626] "Electronic money" means monetary value provided in digital form and used for online transactions and payment of rewards.

[0627] The present invention provides a system that uses a distributed network to achieve high-precision image generation and fairly distributes rewards to terminals that contribute to image generation. Specific embodiments for implementing this system are described below.

[0628] Hardware and Software Use

[0629] In this system, the server uses dedicated software to build and manage the distributed network. Specifically, distributed network software such as Apache Kafka and Ethereum is used. The server has computational resources such as CPU, GPU, and memory, and functions as a system to manage these. The terminals are computers or smart devices that provide the computational resources. The terminals use local resources (CPU / GPU) to process image generation tasks.

[0630] Processing flow

[0631] 1. Network Construction

[0632] The server builds a distributed network and sets up a public IP address to provide an environment in which nodes (terminals) can participate. The server also authenticates nodes and allows them to connect to the network safely.

[0633] 2. Receiving an image generation request

[0634] The user sends a request to the server to generate an image. For example, the user can send a prompt such as "I want you to generate a sunset ocean image."

[0635] 3. Task Decomposition and Allocation

[0636] The server receives an image generation request and divides it into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset." The server distributes these tasks according to the resource availability of each device.

[0637] 4. Task processing and image generation

[0638] The device processes tasks assigned by the server. For example, device A receives the task of "generating sky color" and uses its GPU to generate realistic sky colors. The generated partial images are stored in local storage.

[0639] 5. Image Collection and Evaluation

[0640] The server collects the partial images sent from each device, evaluates them using AI algorithms (such as ResNet or Inception), and selects the best images based on criteria such as texture and color.

[0641] 6. GAN processing and final image generation

[0642] The server generates a highly accurate final image based on the selected images using a Generative Adversarial Network (GAN), which consists of two neural networks, a discriminator and a generator, that learn by competing with each other.

[0643] 7. Reward Allocation

[0644] The server checks the distributed network log and collates which terminal provided how much computing resources. Based on this, it pays the user who provided the terminal a reward in the form of electronic money. The reward is transferred to the user's electronic wallet.

[0645] Specific examples

[0646] Prompt Sentence Examples

[0647] "I want to generate a sunset ocean"

[0648] In this way, our system can efficiently generate highly accurate images and distribute rewards fairly based on the computing resources provided by the terminals.

[0649] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0650] Processing Steps

[0651] Step 1: Building the network

[0652] 1. Specific actions

[0653] To build a distributed network, the server must install and configure dedicated software (e.g., Apache Kafka or Ethereum), set up a public IP address, and create an environment where connections can be made.

[0654] 2. Input

[0655] Necessary network configuration information, software configuration files

[0656] 3. Output

[0657] A fully distributed network, with devices ready to connect

[0658] 4. Detailed explanation

[0659] The server sets up a terminal authentication system to ensure that newly joining terminals can securely connect to the network. To maintain the health of the network, the server monitors the node status in real time.

[0660] Step 2: Receiving an image generation request

[0661] 1. Specific actions

[0662] The user sends a request to the server to generate an image. For example, the user inputs a prompt such as "I want to generate a sunset ocean image."

[0663] 2. Input

[0664] User prompt (e.g. "I want to generate a sunset ocean")

[0665] 3. Output

[0666] Image generation requests recorded in the server's internal database

[0667] 4. Detailed explanation

[0668] The server analyzes the prompt received from the user and records its contents in an internal database. Based on this information, task decomposition is performed in the next processing step.

[0669] Step 3: Break down and assign tasks

[0670] 1. Specific actions

[0671] The server divides the image generation request into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset," and assigns each task to a device.

[0672] 2. Input

[0673] Image generation request from user, device resource information

[0674] 3. Output

[0675] Individual image generation tasks distributed to devices

[0676] 4. Detailed explanation

[0677] The server allocates tasks efficiently, taking into account the resources of each device (CPU, GPU, memory usage), and notifies the device of the allocated tasks.

[0678] Step 4: Receiving and Processing Tasks

[0679] 1. Specific actions

[0680] The device receives tasks assigned by the server and adds them to a local queue. The device executes the tasks using the GPU or CPU to generate images.

[0681] 2. Input

[0682] Image generation tasks assigned by the server

[0683] 3. Output

[0684] Generated partial images, metadata (resources used, processing time, etc.)

[0685] 4. Detailed explanation

[0686] The device uses a local AI model (e.g., CNN or Transformer) based on the task content to generate the specified image elements, and the generated data is stored in local storage.

[0687] Step 5: Collect and evaluate images

[0688] 1. Specific actions

[0689] The server receives the partial images generated by each device and evaluates them using an AI algorithm (e.g., ResNet, Inception).

[0690] 2. Input

[0691] Partial images and metadata sent from the device

[0692] 3. Output

[0693] Evaluation results and selected optimal images

[0694] 4. Detailed explanation

[0695] The server analyzes the texture, color, sharpness, etc. of the collected images and selects the best images based on the evaluation criteria. This selection is used for the next GAN processing.

[0696] Step 6: GAN processing and final image generation

[0697] 1. Specific actions

[0698] The server uses the selected images as input data and generates a highly accurate final image using a Generative Adversarial Network (GAN).

[0699] 2. Input

[0700] Evaluated and selected optimal images

[0701] 3. Output

[0702] High-precision final image

[0703] 4. Detailed explanation

[0704] GAN involves two neural networks, a discriminator and a generator, competing to learn and ultimately generating highly accurate, realistic images, which are then stored in server storage.

[0705] Step 7: Reward Allocation

[0706] 1. Specific actions

[0707] The server checks the distributed network log and calculates the contribution of each device, and based on this, it pays electronic money to the users who provided their devices.

[0708] 2. Input

[0709] Distributed network log, contribution data of each device

[0710] 3. Output

[0711] Rewards sent to the user's electronic wallet

[0712] 4. Detailed explanation

[0713] The server analyzes the usage status of computing resources and processing results of each device, and calculates rewards based on that. The calculated rewards are transferred to the user's electronic wallet as electronic money, realizing fair reward distribution according to the degree of contribution.

[0714] (Application example 1)

[0715] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0716] The problem that the present invention aims to solve is to provide a system that can perform product quality inspections efficiently and with high accuracy in the manufacturing industry. Conventional product inspection methods have limitations in inspection efficiency and accuracy, and require a great deal of effort and cost. Furthermore, conventional methods often involve the subjective judgment of the inspector, making consistent quality control difficult.

[0717] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0718] In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and sending the generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, and means for using the generated images for product inspection and returning the inspection results to the robot. This enables highly accurate and efficient product quality inspection in the manufacturing industry.

[0719] A "distributed network" is a network in which multiple terminals provide computing resources in a distributed manner and work together to process a single task.

[0720] "Computational resources" refer to computer resources such as CPU and GPU performance, memory capacity, and storage capacity provided by a terminal.

[0721] An "image generation task" is a unit of work that includes specific calculations and processing to generate an image.

[0722] "AI" is an abbreviation for artificial intelligence, a technology that uses algorithms such as machine learning and neural networks to analyze data and behave intelligently.

[0723] "Selection" is the process of selecting the best images from the generated images based on evaluation criteria such as texture and color.

[0724] "GAN" is an abbreviation for Generative Adversarial Network, an algorithm in which two neural networks (a generator and a discriminator) compete with each other to learn and generate highly accurate data.

[0725] "Remuneration" is the compensation paid to a terminal that provides computing resources within the system, and is often in the form of electronic money.

[0726] "Product inspection" is the process of checking the quality and condition of manufactured products to ensure they meet standards.

[0727] A "robot" is a device that operates mechanically to perform designated tasks. It is often used in automated manufacturing processes.

[0728] MODE FOR CARRYING OUT THE INVENTION

[0729] As an embodiment of the present invention, a system for performing high-precision product inspection in a factory using a distributed network will be described. This system operates mainly in cooperation between a server, a terminal (robot), and a user.

[0730] Server Features

[0731] 1. Networking:

[0732] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[0733] 2. Task distribution:

[0734] The server receives product inspection requests from users and breaks them down into smaller tasks, such as inspecting specific parts or checking specific details. It then distributes these tasks to each terminal and manages them appropriately.

[0735] 3. Image Collection and Analysis:

[0736] The server collects product images acquired from each device, analyzes and evaluates the collected images using AI algorithms, and selects the most suitable images based on criteria such as the presence or absence of defects.

[0737] 4. GAN Processing:

[0738] The server uses GAN (Generic Annotation) to generate a highly accurate final image based on the selected images. GAN consists of two neural networks, a discriminator and a generator, that compete with each other to learn and provide more accurate inspection results.

[0739] 5. Reward Allocation:

[0740] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0741] 6. Return of test results:

[0742] The server compiles the test results and sends the final test results back to the robot.

[0743] Terminal (robot) functions

[0744] 1. Registering with the Network:

[0745] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0746] 2. Receiving and processing tasks:

[0747] The terminal receives and processes image generation and inspection tasks assigned to it by the server, performing specific tasks such as capturing images to inspect a specific part or evaluating the generated images to identify specific details.

[0748] 3. Sending images:

[0749] The device captures an image of the product and sends it to a server, along with metadata such as computing resource usage.

[0750] User Roles

[0751] 1. Product Inspection Request:

[0752] The user requests a specific product inspection from the server, specifying, for example, "Please inspect the surface of the part for scratches or dirt" or "Please check the product for color unevenness or shape abnormalities."

[0753] 2. Receiving rewards:

[0754] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0755] Hardware and software used

[0756] Hardware:

[0757] A robot installed on a factory production line (equipped with an image capture camera)

[0758] Modem / router for network connection

[0759] software:

[0760] Implemented server and robot control programs using Python.

[0761] Use the Requests library to process HTTP requests.

[0762] Use a local server or a cloud server (e.g. AWS, Google Cloud) on the server side.

[0763] Examples:

[0764] For example, one product inspection request might be "surface scratch inspection." The server splits this request into two tasks: "capturing an image of a specific part" and "analyzing the captured image using AI." The captured image is sent to the server and analyzed by an AI algorithm. The analysis results are used to determine whether there are any surface scratches using a deep learning model, and the results are sent back to the robot. The robot then decides on the next action to take on the production line.

[0765] Example prompt: "Inspect the product surface for scratches or stains."

[0766] This system enables factories to automate product inspection with high precision and efficiency, dramatically improving quality control.

[0767] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0768] Step 1: Register with the network

[0769] Server: The server sets a public IP address and builds a distributed network. It prepares the network environment so that devices can join. The input is the public IP address and network setting information, and the output is a network environment that devices can connect to.

[0770] Device: The device accesses the server to register on the network. As registration information, it sends its device ID and performance (CPU / GPU specs). The input is the device ID and performance information, and the output is a network participation token issued by the server.

[0771] Step 2: Receiving and processing the image generation task

[0772] Server: The server receives product inspection requests from users and divides them into image generation tasks. Tasks include inspecting specific parts and checking details. The input is the inspection request from the user, and the output is the image generation tasks assigned to each terminal.

[0773] Terminal: The terminal processes tasks received from the server. For example, it takes an image of a specific part of a product and sends the captured data to the server. The input is the image generation task from the server, and the output is the captured product image.

[0774] Step 3: Submit your images and add metadata

[0775] Terminal: When generating product images, metadata such as the usage of computing resources is added to the images when they are sent to the server. The input is the captured product image and metadata, and the output is a data package sent to the server.

[0776] Server: The server stores the received images and metadata and proceeds to the next analysis step. The input is the data package sent from the device, and the output is the stored image data and metadata.

[0777] Step 4: Analyze and select images

[0778] Server: The server uses an AI algorithm to analyze the received images. It selects the most suitable images based on criteria such as texture and color. The input is the stored image data and metadata, and the output is the selected image.

[0779] Step 5: GAN Processing

[0780] Server: The server generates a highly accurate final image using a generative anti-theft network (GAN) based on the filtered image. The input is the filtered image, and the output is the highly accurate final image.

[0781] Step 6: Generate and return test results

[0782] Server: Generates product inspection results based on the final image and sends the results back to the terminal. The input is the final image generated with high accuracy, and the output is the inspection result data.

[0783] Terminal: The terminal processes the inspection results received from the server and executes the next instructions on the production line. The input is the inspection result data from the server, and the output is the instructions for the production line.

[0784] Step 7: Compensation and Allocation

[0785] Server: The server checks the distributed network log and calculates rewards based on the provision of computing resources by each terminal. The input is the network log and computing resource information, and the output is the reward amount for each terminal.

[0786] User: The user who provides the terminal receives the reward in the form of electronic money. The input is the reward information from the server, and the output is a transfer to the user's electronic wallet.

[0787] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0788] As an embodiment of the present invention, we will explain a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. The system mainly operates in cooperation between a server, terminals, and users.

[0789] Server Features

[0790] 1. Network Construction

[0791] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[0792] 2. Task Distribution

[0793] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[0794] 3. Image collection and analysis

[0795] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[0796] 4. Emotional Engine Adjustment

[0797] The server uses an emotion engine to analyze the user's emotions in real time and adjusts image generation parameters and task distribution based on the analysis.

[0798] 5. GAN Processing

[0799] The server then uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs use two neural networks, a discriminator and a generator, to compete with each other to learn and generate more realistic images.

[0800] 6. Reward Allocation

[0801] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[0802] Device Features

[0803] 1. Registering with the Network

[0804] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[0805] 2. Receiving and processing tasks

[0806] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[0807] 3. Sending images

[0808] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[0809] User Roles

[0810] 1. Request for image generation

[0811] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[0812] 2. Providing emotions

[0813] Users use cameras and sensors to send facial and voice data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[0814] 3. Receiving rewards

[0815] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[0816] Specific examples

[0817] Image generation request example

[0818] 1. Receiving a request

[0819] The server receives a request from the user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[0820] 2. Task Distribution and Processing

[0821] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[0822] 3. Image selection and GAN processing

[0823] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0824] 4. Emotional Engine Adjustment

[0825] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. Based on this data, the server adjusts the parameters required for the next image generation.

[0826] 5. Reward Allocation

[0827] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0828] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[0829] The processing flow will be explained below.

[0830] Step 1:

[0831] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[0832] Step 2:

[0833] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[0834] Step 3:

[0835] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[0836] Step 4:

[0837] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[0838] Step 5:

[0839] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[0840] Step 6:

[0841] Terminal A and terminal B process the received task and generate images. Terminal A generates an image with a sand texture, and terminal B generates an image with a mountain silhouette.

[0842] Step 7:

[0843] Once the generated images are complete, device A and device B send each generated image to the server, along with metadata such as the usage of computing resources.

[0844] Step 8:

[0845] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[0846] Step 9:

[0847] The server uses GAN to generate a highly accurate final image based on the selected images. In this process, two neural networks, a discriminator and a generator, compete with each other to learn and generate realistic images.

[0848] Step 10:

[0849] While the user evaluates the generated images, real-time emotional data is collected via a camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction.

[0850] Step 11:

[0851] The server adjusts the parameters required for the next image generation based on the data analyzed by the emotion engine. For example, if the user is not satisfied, it adjusts the color tone or details.

[0852] Step 12:

[0853] The server checks the distributed network logs and records which devices provided how much computing resources.

[0854] Step 13:

[0855] The server calculates rewards based on the amount of computing resources provided. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[0856] Step 14:

[0857] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[0858] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and has an adjustment function that reflects the user's emotional data in real time, ultimately achieving appropriate reward distribution.

[0859] Example 2

[0860] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0861] Conventional image generation systems lack the technology to rapidly generate highly accurate images or to generate images that reflect user emotions. They also lack the means to efficiently utilize distributed networks and distribute appropriate rewards based on computational resources. To address these issues, a system that integrates advanced image analysis and generation technology with the ability to reflect users' real-time emotions is needed.

[0862] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0863] In this invention, the server includes a means for constructing a distributed network, a means for supplying computing resources to each terminal, and a means for distributing image generation tasks. This enables highly accurate image generation and further enables real-time analysis of user emotions and their reflection in image generation. It also makes it possible to allocate appropriate rewards to each terminal based on the provision of computing resources.

[0864] A "distributed network" is a network structure in which multiple terminals work together to process data and communicate without relying on a centralized server.

[0865] "Computational resources" refers to the hardware and performance required for data processing and calculations, such as a computer system's CPU, GPU, memory, and storage.

[0866] An "image generation task" is a specific task or instruction for generating a new image using a particular algorithm or process.

[0867] "AI analysis and selection" means using artificial intelligence technology to analyze input data or information and select the best option based on specific criteria.

[0868] An "emotion engine" is a technology that analyzes emotions from a user's facial expressions and voice and adjusts the system's operation based on the results.

[0869] "GAN" stands for Generative Adversarial Network, a technology in which two neural networks, a generator and a discriminator, learn by competing with each other to generate highly accurate data (such as images).

[0870] "Electronic money" is digital or virtual currency that has value in electronic form and is traded over the Internet, rather than physical cash.

[0871] A "log" is data that represents a record of system actions or events, which can later be used for analysis or auditing.

[0872] This invention is a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. This system operates mainly through the cooperation of three elements: the server, the terminal, and the user. The roles and specific operations of each are described below.

[0873] Server Features

[0874] 1. Network Construction

[0875] The server creates a distributed network and sets up a public IP address for devices to join, allowing them to contact the server and register on the network.

[0876] 2. Task Distribution

[0877] The server receives an image generation request from a user, divides the request into small tasks, and distributes them to each terminal. At this time, it assigns the task to the most suitable terminal depending on the type and difficulty of the task.

[0878] 3. Image collection and analysis

[0879] The server collects the images generated by each device and uses AI algorithms to analyze and evaluate them, including texture and color, to select the best images.

[0880] 4. Emotional Engine Adjustment

[0881] The server uses an emotion engine to analyze the user's emotion data in real time and adjusts the image generation parameters based on the results, making it possible to generate images that are in line with the user's emotions.

[0882] 5. GAN Processing

[0883] The server uses a generative adversarial network (GAN) to generate a highly accurate final image based on the selected images. GANs are made up of two neural networks, a generator and a discriminator, that compete with each other to learn and generate more realistic images.

[0884] 6. Reward Allocation

[0885] The server checks the distributed network log and pays each terminal electronic money in return for the computing resources provided.

[0886] Device Features

[0887] 1. Registering with the Network

[0888] The device accesses the server and registers its computing resources (CPU / GPU performance, etc.) After registration, the device operates as part of the distributed network and is ready to receive tasks.

[0889] 2. Receiving and processing tasks

[0890] The device receives image generation tasks assigned by the server and performs the tasks, such as "generating the texture of sand" or "generating a night view of a city."

[0891] 3. Sending images

[0892] When the generated image is sent to the server, metadata such as the usage status of computing resources is also sent.

[0893] User Roles

[0894] 1. Request for image generation

[0895] A user requests the server to generate a specific image, specifically, by describing the request, such as a "mountain landscape" or a "city night view."

[0896] 2. Providing emotions

[0897] Users use cameras and sensors to send their emotional data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[0898] 3. Receiving rewards

[0899] Users who provide their devices receive rewards based on the tasks their devices complete, and the rewards are transferred to the users' electronic wallets as electronic money.

[0900] Specific examples

[0901] Image generation request example

[0902] 1. Receiving a request

[0903] The server receives a request from a user, such as "I want you to generate a desert landscape." The request is divided into smaller tasks, such as "generate the texture of sand" and "generate distant mountains."

[0904] 2. Task Distribution and Processing

[0905] The server assigns the task of "generating a sand texture" to terminal A and the task of "generating a silhouette of a distant mountain" to terminal B. Terminal A generates the sand texture, and terminal B generates the mountain silhouette, and they each send them to the server.

[0906] 3. Image selection and GAN processing

[0907] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[0908] 4. Emotional Engine Adjustment

[0909] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. The server then adjusts the parameters required for the next image generation based on this data.

[0910] 5. Reward Allocation

[0911] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[0912] Prompt Sentence Examples

[0913] "I want a desert landscape to be generated."

[0914] "I want to see the city's night view."

[0915] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[0916] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0917] Step 1:

[0918] Server: Network construction

[0919] The server sets up a distributed network by setting up a public IP address, which allows devices to join the network. The input requires the server's network configuration information, and the output is an initialized network. Specifically, the server edits the network configuration file, defines the public IP address and port number, and then listens on the network.

[0920] Step 2:

[0921] End user: Input of image generation request

[0922] A user requests a server to generate a specific image. The input is the user's request (e.g., "desert landscape"), and the output is the request being processed by the server. Specifically, the user enters the statement "I want a desert landscape generated" into a form in a web application and presses the submit button.

[0923] Step 3:

[0924] Server: Analyzes the request and distributes tasks

[0925] The server analyzes the received request and divides the task into smaller parts. The input is the user's request, and the output is the divided tasks. Specifically, the server divides the task into tasks such as "generate the texture of sand" and "generate the silhouette of a distant mountain," and distributes each task to the terminal.

[0926] Step 4:

[0927] Terminal: Receiving and processing tasks

[0928] The device receives tasks from the server and generates images. The input is the task content distributed by the server (e.g., "generate the texture of sand"), and the output is the generated image. Specifically, the device uses a deep learning model to generate the texture of sand and create its image data.

[0929] Step 5:

[0930] Device: Send image

[0931] The device sends the generated image to the server. The input is the generated image data, and the output is the image data sent to the server. Specifically, the device uploads the generated image data to the API endpoint specified by the server.

[0932] Step 6:

[0933] Server: Image collection and analysis

[0934] The server collects images received from the devices and analyzes them using an AI algorithm. The input is the image data sent from each device, and the output is the evaluated image data. Specifically, the server selects the most suitable images based on evaluation criteria such as the texture and color of each image.

[0935] Step 7:

[0936] Server: Coordination by emotion engine

[0937] The server analyzes the user's emotional data using an emotion engine and adjusts the image generation parameters. The input is the user's emotional data (for example, facial expression data and voice data), and the output is the adjusted parameters. Specifically, the server collects the user's emotional data, analyzes it using the emotion engine, and adjusts the parameters of the image generation algorithm based on the results.

[0938] Step 8:

[0939] Server: High-precision image generation using GAN processing

[0940] The server generates a high-precision image using a GAN model based on the selected images. The input is the evaluated image data, and the output is the final high-precision image. Specifically, the server inputs the selected image data into the generator, and has it compete with the discriminator to progress learning, ultimately generating a high-precision image.

[0941] Step 9:

[0942] Server: Reward Distribution

[0943] The server checks the distributed network log and calculates rewards based on the provision of computing resources. The input is the processing log data of each terminal, and the output is the reward amount for each terminal. Specifically, the server analyzes the processing log of each terminal and transfers the calculated electronic money as a reward to the electronic wallet of each terminal owner.

[0944] (Application example 2)

[0945] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0946] Conventional image generation systems require large amounts of computational resources to generate highly accurate images, and they have difficulty reflecting the specific emotional satisfaction desired by users in real time. Furthermore, there is a lack of a way to effectively utilize user feedback in the image generation process. As a result, images that do not fully meet the user's requirements are often generated, and satisfactory results cannot be achieved.

[0947] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and transmitting generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, means for collecting real-time emotional data from users, analyzing it, and reflecting it in the image generation process, and means for linking images generated via a terminal such as a smartphone with user emotional feedback. This makes it possible to generate high-precision images while also reflecting user emotional feedback in real time, thereby achieving image generation that greatly improves user satisfaction.

[0948] A "distributed network" is a network in which multiple devices are interconnected and can share and process data without centralized control.

[0949] "Computational resources" refers to the capabilities of hardware and software to perform computational processing, and specifically includes the performance of CPUs and GPUs.

[0950] "Image generation task" means a set of instructions for processing or computing required to generate a particular image.

[0951] A "server" is a central computer that processes and provides data in response to requests from multiple terminals within a network.

[0952] A "terminal" refers to a computer or device that is connected to a distributed network and provides computing resources.

[0953] "AI analysis and selection methods" refers to technology that uses artificial intelligence to evaluate collected images and select the most suitable ones.

[0954] "GAN" stands for generative adversarial network, a type of algorithm that generates realistic images by training two neural networks, a generator and a discriminator, to compete with each other.

[0955] "High-precision images" refer to detailed, high-quality images that closely resemble real-world scenes and objects.

[0956] "Electronic money" means currency provided in digital form, an asset that is exchanged electronically.

[0957] "Real-time emotional data" refers to information that collects the user's emotional state in real time and processes that data immediately.

[0958] "Emotional feedback" refers to the evaluation and response of a system based on emotional expressions and data provided by the user.

[0959] This invention is a system that uses a distributed network to generate highly accurate images, and also has the function of recognizing and reflecting user emotions in real time. The following is a specific embodiment of this invention.

[0960] Server Features

[0961] The server has the following main functions:

[0962] 1. Networking:

[0963] The server creates a distributed network and configures it so that each terminal can provide computing resources. It uses public IP addresses to provide an environment where terminals can easily connect.

[0964] 2. Task distribution:

[0965] The system divides image generation requests received from users into small tasks and distributes them to each device. For example, it gives detailed instructions such as "generate the texture of sand" or "generate the silhouette of a distant mountain" to each device.

[0966] 3. Image Collection and Analysis:

[0967] Images generated by each device are collected and analyzed and evaluated using AI algorithms, and the best images are selected based on evaluation criteria such as texture, color, and composition.

[0968] 4. Emotional engine regulation:

[0969] The system analyzes the user's emotional data in real time and reflects it at each stage of the generation process. This is achieved by analyzing the emotional information provided by the user through a camera or microphone.

[0970] 5. GAN Processing:

[0971] To generate the final high-precision image, a generative adversarial network (GAN) is used based on the selected images. GAN includes two neural networks: a generator and a discriminator.

[0972] 6. Reward Allocation:

[0973] The system checks the distributed network logs, calculates rewards based on the computing resources provided, and pays each device electronic money.

[0974] Device Features

[0975] The terminal has the following features:

[0976] 1. Registering with the Network:

[0977] They connect to a server, register with the decentralized network, provide computing resources, and receive tokens in return.

[0978] 2. Receiving and processing tasks:

[0979] It receives and processes image generation tasks from the server, and includes metadata such as the usage of computing resources when sending the generated images to the server.

[0980] User Roles

[0981] Users have the following roles:

[0982] 1. Image generation request:

[0983] Request the server to generate a specific image, specifying specific content such as a "mountain landscape" or a "city night view."

[0984] 2. Providing Emotion Data:

[0985] Cameras and sensors are used to provide facial and voice data in real time, which is analyzed by the emotion engine and reflected in image generation.

[0986] Specific examples

[0987] A user opens the app and requests that an image of a "summer beach sunset" be generated. The app collects the user's emotional data in real time via the camera and microphone. When the user expresses emotions such as "happy" or "moved" while looking at the image, that data is sent to the server and reflected in the image generation. Finally, an image that satisfies the user is generated and displayed on the smartphone.

[0988] An example prompt for a generative AI model is:

[0989] "Generate a summer seaside scene. I'd like the image to have a tranquil, relaxing atmosphere with the sound of the waves. Additionally, I'd like the sunset to be reflected on the water to enhance its beauty."

[0990] The above is an embodiment of the present invention. This system not only generates highly accurate images but also reflects real-time emotional feedback from the user, thereby improving user satisfaction.

[0991] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0992] Step 1:

[0993] Request Receipt

[0994] The server receives an image generation request from a user.

[0995] (Input) The user's request (e.g., "Summer sunset at the beach").

[0996] (Output) Initial settings for the image generation task.

[0997] (Specific operation) A user requests the generation of a specific image through the app, and the server receives the request, which may include a prompt for the generating AI model.

[0998] Step 2:

[0999] Task division

[1000] The server divides the image generation request into multiple subtasks.

[1001] (Input) The received image generation request.

[1002] (Output) A set of divided image generation tasks.

[1003] (Specific operation) The server breaks down the data into specific parts such as "generating the texture of sand" or "adjusting the color of the sunset" and converts each part into a processable format.

[1004] Step 3:

[1005] Task Distribution

[1006] The server distributes the divided tasks to the appropriate terminals.

[1007] (Input) A set of divided image generation tasks.

[1008] (Output) Tasks sent to each device.

[1009] (Specific operation) The server assigns tasks and sends instructions based on each terminal's available computing resources and past performance.

[1010] Step 4:

[1011] Receiving and processing tasks

[1012] The terminal processes the received task and sends the results to the server.

[1013] (Input) Image generation task sent from the server.

[1014] (Output) The generated subimage and metadata.

[1015] (Specific operation) Each device executes the received task, specifically generating the texture and color of a specific image, and sends this to the server along with metadata including the usage status of computing resources.

[1016] Step 5:

[1017] Image collection and analysis

[1018] The server collects images sent from each device and analyzes them using AI algorithms.

[1019] (Input) Partial images and metadata sent from each device.

[1020] (Output) The analyzed image data.

[1021] (Specific operation) The server evaluates the texture, color, composition, etc. of each image and selects the most suitable image using an AI algorithm.

[1022] Step 6:

[1023] Emotion data collection and analysis

[1024] The server collects and analyzes the user's real-time emotional data.

[1025] (Input) Emotion data from the user via camera and microphone.

[1026] (Output) Parsed emotion information.

[1027] (Specific operation) While the user is checking the generated image, facial expression and voice data is collected via the smartphone's camera and microphone, and the server analyzes this data using the emotion engine.

[1028] Step 7:

[1029] Reflecting and adjusting emotional data

[1030] The server adjusts image generation parameters based on the analyzed emotional information.

[1031] (Input) Analyzed emotion information.

[1032] (Output) Adjusted image generation parameters.

[1033] (Specific operation) The server adjusts the parameters of the generation process according to the user's positive or negative emotions. This adjustment ensures that the next task executed reflects the emotional data.

[1034] Step 8:

[1035] Final GAN ​​processing

[1036] The server generates the final image using GAN based on the selected images.

[1037] (Input) Selected partial images and adjusted parameters.

[1038] (Output) High-precision final image.

[1039] (Specific operation) Using a GAN (generative adversarial network) with a generator and discriminator, the final result is a highly accurate and realistic image.

[1040] Step 9:

[1041] Final image courtesy

[1042] The server provides the generated final image to the user.

[1043] (Input) High-precision final image.

[1044] (Output) The image that is displayed on the user device.

[1045] (Specific operation) The server sends the generated final image to the user's device, and the app displays it.

[1046] Step 10:

[1047] Reward Allocation

[1048] The server distributes rewards to the terminals that participate in the task processing.

[1049] (Input) Distributed network log and contribution data of each device.

[1050] (Output) Reward payment to each device.

[1051] (Specific operation) The server analyzes the logs of the distributed network, calculates rewards based on the degree of contribution, and pays electronic money to each terminal.

[1052] This concludes the explanation of the specific processing steps. This system enables highly accurate image generation and reflects the user's emotional feedback in real time.

[1053] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1054] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1055] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1056] [Third embodiment]

[1057] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1058] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1059] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1060] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1061] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1062] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1063] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1064] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1065] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1066] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1067] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1068] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1069] As an embodiment of the present invention, a system for generating highly accurate images using a distributed network and paying rewards to terminals that contribute to the image generation will be described below. The system operates mainly in cooperation between a server, terminals, and users.

[1070] Server Features

[1071] 1. Network Construction

[1072] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[1073] 2. Task Distribution

[1074] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[1075] 3. Image collection and analysis

[1076] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[1077] 4. GAN Processing

[1078] The server uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs are made up of two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[1079] 5. Reward Allocation

[1080] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1081] Device Features

[1082] 1. Registering with the Network

[1083] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1084] 2. Receiving and processing tasks

[1085] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[1086] 3. Sending images

[1087] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[1088] User Roles

[1089] 1. Request for image generation

[1090] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[1091] 2. Receiving Rewards

[1092] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1093] Specific examples

[1094] Image generation request example

[1095] 1. Receiving a request

[1096] The server receives a request from a user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[1097] 2. Task Distribution and Processing

[1098] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[1099] 3. Image selection and GAN processing

[1100] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1101] 4. Reward Allocation

[1102] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1103] The above is a detailed description of the embodiment of the present invention. This system realizes efficient and highly accurate image generation and appropriate reward distribution using a distributed network.

[1104] The processing flow will be explained below.

[1105] Step 1:

[1106] The server creates a distributed network and configures it so that nodes (devices) can join, which includes setting up public IP addresses so that devices can connect to the network.

[1107] Step 2:

[1108] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[1109] Step 3:

[1110] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[1111] Step 4:

[1112] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[1113] Step 5:

[1114] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[1115] Step 6:

[1116] The devices process the received tasks and generate images. Device A generates an image with a sandy texture, while device B generates an image with a mountain silhouette.

[1117] Step 7:

[1118] The device sends the generated images to the server, and metadata such as computational resource usage is included with each image.

[1119] Step 8:

[1120] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[1121] Step 9:

[1122] The server then uses GAN to generate a highly accurate final image based on the selected images. GAN uses two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[1123] Step 10:

[1124] The server checks the distributed network logs and records which devices provided how much computing resources.

[1125] Step 11:

[1126] The server calculates rewards based on the provision of computing resources. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[1127] Step 12:

[1128] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[1129] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and achieves appropriate reward distribution.

[1130] Example 1

[1131] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1132] This invention relates to a system that efficiently realizes high-precision image generation and fairly distributes rewards to terminals that provide computational resources that contribute to image generation. Conventional systems mainly use a centralized approach, which has led to problems such as the concentration of computational resources and unfair distribution of rewards. Furthermore, high-precision image generation requires a large amount of computational resources, which calls for a distributed computing environment.

[1133] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1134] In this invention, the server includes a means for constructing a distributed network, a means for providing computational resources to each terminal, and a means for receiving and decomposing image generation requests. This enables efficient high-precision image generation and fair distribution of rewards to terminals that provide computational resources. Specifically, the server assigns image generation tasks to each terminal, evaluates the generated images, and generates a high-precision final image. The server also checks the distributed network log, calculates rewards based on the provision of computational resources, and pays them in electronic money, thereby achieving overall system efficiency and fairness.

[1135] A "distributed network" is a network in which terminals with multiple computing resources are connected on an equal basis and perform data processing and communication without central control.

[1136] "Computational resources" is a general term for hardware resources used for data processing and calculations, such as a terminal's CPU, GPU, and memory.

[1137] An "image generation request" is a request or prompt that a user sends to a server to generate a particular image.

[1138] An "image generation task" is a small unit of work required to realize an image generation request, and is distributed to each terminal.

[1139] "Generative network" refers to a generative adversarial network (GAN), an algorithm in which two neural networks, a discriminator and a generator, compete to generate data.

[1140] "Task allocation" refers to the operations and processes for appropriately distributing image generation tasks to each terminal.

[1141] "Reward" refers to the compensation paid to the terminal that contributed to the image generation, and is usually provided in the form of electronic money.

[1142] "Electronic money" means monetary value provided in digital form and used for online transactions and payment of rewards.

[1143] The present invention provides a system that uses a distributed network to achieve high-precision image generation and fairly distributes rewards to terminals that contribute to image generation. Specific embodiments for implementing this system are described below.

[1144] Hardware and Software Use

[1145] In this system, the server uses dedicated software to build and manage the distributed network. Specifically, distributed network software such as Apache Kafka and Ethereum is used. The server has computational resources such as CPU, GPU, and memory, and functions as a system to manage these. The terminals are computers or smart devices that provide the computational resources. The terminals use local resources (CPU / GPU) to process image generation tasks.

[1146] Processing flow

[1147] 1. Network Construction

[1148] The server builds a distributed network and sets up a public IP address to provide an environment in which nodes (terminals) can participate. The server also authenticates nodes and allows them to connect to the network safely.

[1149] 2. Receiving an image generation request

[1150] The user sends a request to the server to generate an image. For example, the user can send a prompt such as "I want you to generate a sunset ocean image."

[1151] 3. Task Decomposition and Allocation

[1152] The server receives an image generation request and divides it into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset." The server distributes these tasks according to the resource availability of each device.

[1153] 4. Task processing and image generation

[1154] The device processes tasks assigned by the server. For example, device A receives the task of "generating sky color" and uses its GPU to generate realistic sky colors. The generated partial images are stored in local storage.

[1155] 5. Image Collection and Evaluation

[1156] The server collects the partial images sent from each device, evaluates them using AI algorithms (such as ResNet or Inception), and selects the best images based on criteria such as texture and color.

[1157] 6. GAN processing and final image generation

[1158] The server generates a highly accurate final image based on the selected images using a Generative Adversarial Network (GAN), which consists of two neural networks, a discriminator and a generator, that learn by competing with each other.

[1159] 7. Reward Allocation

[1160] The server checks the distributed network log and collates which terminal provided how much computing resources. Based on this, it pays the user who provided the terminal a reward in the form of electronic money. The reward is transferred to the user's electronic wallet.

[1161] Specific examples

[1162] Prompt Sentence Examples

[1163] "I want to generate a sunset ocean"

[1164] In this way, our system can efficiently generate highly accurate images and distribute rewards fairly based on the computing resources provided by the terminals.

[1165] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1166] Processing Steps

[1167] Step 1: Building the network

[1168] 1. Specific actions

[1169] To build a distributed network, the server must install and configure dedicated software (e.g., Apache Kafka or Ethereum), set up a public IP address, and create an environment where connections can be made.

[1170] 2. Input

[1171] Necessary network configuration information, software configuration files

[1172] 3. Output

[1173] A fully distributed network, with devices ready to connect

[1174] 4. Detailed explanation

[1175] The server sets up a terminal authentication system to ensure that newly joining terminals can securely connect to the network. To maintain the health of the network, the server monitors the node status in real time.

[1176] Step 2: Receiving an image generation request

[1177] 1. Specific actions

[1178] The user sends a request to the server to generate an image. For example, the user inputs a prompt such as "I want to generate a sunset ocean image."

[1179] 2. Input

[1180] User prompt (e.g. "I want to generate a sunset ocean")

[1181] 3. Output

[1182] Image generation requests recorded in the server's internal database

[1183] 4. Detailed explanation

[1184] The server analyzes the prompt received from the user and records its contents in an internal database. Based on this information, task decomposition is performed in the next processing step.

[1185] Step 3: Break down and assign tasks

[1186] 1. Specific actions

[1187] The server divides the image generation request into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset," and assigns each task to a device.

[1188] 2. Input

[1189] Image generation request from user, device resource information

[1190] 3. Output

[1191] Individual image generation tasks distributed to devices

[1192] 4. Detailed explanation

[1193] The server allocates tasks efficiently, taking into account the resources of each device (CPU, GPU, memory usage), and notifies the device of the allocated tasks.

[1194] Step 4: Receiving and Processing Tasks

[1195] 1. Specific actions

[1196] The device receives tasks assigned by the server and adds them to a local queue. The device executes the tasks using the GPU or CPU to generate images.

[1197] 2. Input

[1198] Image generation tasks assigned by the server

[1199] 3. Output

[1200] Generated partial images, metadata (resources used, processing time, etc.)

[1201] 4. Detailed explanation

[1202] The device uses a local AI model (e.g., CNN or Transformer) based on the task content to generate the specified image elements, and the generated data is stored in local storage.

[1203] Step 5: Collect and evaluate images

[1204] 1. Specific actions

[1205] The server receives the partial images generated by each device and evaluates them using an AI algorithm (e.g., ResNet, Inception).

[1206] 2. Input

[1207] Partial images and metadata sent from the device

[1208] 3. Output

[1209] Evaluation results and selected optimal images

[1210] 4. Detailed explanation

[1211] The server analyzes the texture, color, sharpness, etc. of the collected images and selects the best images based on the evaluation criteria. This selection is used for the next GAN processing.

[1212] Step 6: GAN processing and final image generation

[1213] 1. Specific actions

[1214] The server uses the selected images as input data and generates a highly accurate final image using a Generative Adversarial Network (GAN).

[1215] 2. Input

[1216] Evaluated and selected optimal images

[1217] 3. Output

[1218] High-precision final image

[1219] 4. Detailed explanation

[1220] GAN involves two neural networks, a discriminator and a generator, competing to learn and ultimately generating highly accurate, realistic images, which are then stored in server storage.

[1221] Step 7: Reward Allocation

[1222] 1. Specific actions

[1223] The server checks the distributed network log and calculates the contribution of each device, and based on this, it pays electronic money to the users who provided their devices.

[1224] 2. Input

[1225] Distributed network log, contribution data of each device

[1226] 3. Output

[1227] Rewards sent to the user's electronic wallet

[1228] 4. Detailed explanation

[1229] The server analyzes the usage status of computing resources and processing results of each device, and calculates rewards based on that. The calculated rewards are transferred to the user's electronic wallet as electronic money, realizing fair reward distribution according to the degree of contribution.

[1230] (Application example 1)

[1231] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1232] The problem that the present invention aims to solve is to provide a system that can perform product quality inspections efficiently and with high accuracy in the manufacturing industry. Conventional product inspection methods have limitations in inspection efficiency and accuracy, and require a great deal of effort and cost. Furthermore, conventional methods often involve the subjective judgment of the inspector, making consistent quality control difficult.

[1233] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1234] In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and sending the generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, and means for using the generated images for product inspection and returning the inspection results to the robot. This enables highly accurate and efficient product quality inspection in the manufacturing industry.

[1235] A "distributed network" is a network in which multiple terminals provide computing resources in a distributed manner and work together to process a single task.

[1236] "Computational resources" refer to computer resources such as CPU and GPU performance, memory capacity, and storage capacity provided by a terminal.

[1237] An "image generation task" is a unit of work that includes specific calculations and processing to generate an image.

[1238] "AI" is an abbreviation for artificial intelligence, a technology that uses algorithms such as machine learning and neural networks to analyze data and behave intelligently.

[1239] "Selection" is the process of selecting the best images from the generated images based on evaluation criteria such as texture and color.

[1240] "GAN" is an abbreviation for Generative Adversarial Network, an algorithm in which two neural networks (a generator and a discriminator) compete with each other to learn and generate highly accurate data.

[1241] "Remuneration" is the compensation paid to a terminal that provides computing resources within the system, and is often in the form of electronic money.

[1242] "Product inspection" is the process of checking the quality and condition of manufactured products to ensure they meet standards.

[1243] A "robot" is a device that operates mechanically to perform designated tasks. It is often used in automated manufacturing processes.

[1244] MODE FOR CARRYING OUT THE INVENTION

[1245] As an embodiment of the present invention, a system for performing high-precision product inspection in a factory using a distributed network will be described. This system operates mainly in cooperation between a server, a terminal (robot), and a user.

[1246] Server Features

[1247] 1. Networking:

[1248] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[1249] 2. Task distribution:

[1250] The server receives product inspection requests from users and breaks them down into smaller tasks, such as inspecting specific parts or checking specific details. It then distributes these tasks to each terminal and manages them appropriately.

[1251] 3. Image Collection and Analysis:

[1252] The server collects product images acquired from each device, analyzes and evaluates the collected images using AI algorithms, and selects the most suitable images based on criteria such as the presence or absence of defects.

[1253] 4. GAN Processing:

[1254] The server uses GAN (Generic Annotation) to generate a highly accurate final image based on the selected images. GAN consists of two neural networks, a discriminator and a generator, that compete with each other to learn and provide more accurate inspection results.

[1255] 5. Reward Allocation:

[1256] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1257] 6. Return of test results:

[1258] The server compiles the test results and sends the final test results back to the robot.

[1259] Terminal (robot) functions

[1260] 1. Registering with the Network:

[1261] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1262] 2. Receiving and processing tasks:

[1263] The terminal receives and processes image generation and inspection tasks assigned to it by the server, performing specific tasks such as capturing images to inspect a specific part or evaluating the generated images to identify specific details.

[1264] 3. Sending images:

[1265] The device captures an image of the product and sends it to a server, along with metadata such as computing resource usage.

[1266] User Roles

[1267] 1. Product Inspection Request:

[1268] The user requests a specific product inspection from the server, specifying, for example, "Please inspect the surface of the part for scratches or dirt" or "Please check the product for color unevenness or shape abnormalities."

[1269] 2. Receiving rewards:

[1270] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1271] Hardware and software used

[1272] Hardware:

[1273] A robot installed on a factory production line (equipped with an image capture camera)

[1274] Modem / router for network connection

[1275] software:

[1276] Implemented server and robot control programs using Python.

[1277] Use the Requests library to process HTTP requests.

[1278] Use a local server or a cloud server (e.g. AWS, Google Cloud) on the server side.

[1279] Examples:

[1280] For example, one product inspection request might be "surface scratch inspection." The server splits this request into two tasks: "capturing an image of a specific part" and "analyzing the captured image using AI." The captured image is sent to the server and analyzed by an AI algorithm. The analysis results are used to determine whether there are any surface scratches using a deep learning model, and the results are sent back to the robot. The robot then decides on the next action to take on the production line.

[1281] Example prompt: "Inspect the product surface for scratches or stains."

[1282] This system enables factories to automate product inspection with high precision and efficiency, dramatically improving quality control.

[1283] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1284] Step 1: Register with the network

[1285] Server: The server sets a public IP address and builds a distributed network. It prepares the network environment so that devices can join. The input is the public IP address and network setting information, and the output is a network environment that devices can connect to.

[1286] Device: The device accesses the server to register on the network. As registration information, it sends its device ID and performance (CPU / GPU specs). The input is the device ID and performance information, and the output is a network participation token issued by the server.

[1287] Step 2: Receiving and processing the image generation task

[1288] Server: The server receives product inspection requests from users and divides them into image generation tasks. Tasks include inspecting specific parts and checking details. The input is the inspection request from the user, and the output is the image generation tasks assigned to each terminal.

[1289] Terminal: The terminal processes tasks received from the server. For example, it takes an image of a specific part of a product and sends the captured data to the server. The input is the image generation task from the server, and the output is the captured product image.

[1290] Step 3: Submit your images and add metadata

[1291] Terminal: When generating product images, metadata such as the usage of computing resources is added to the images when they are sent to the server. The input is the captured product image and metadata, and the output is a data package sent to the server.

[1292] Server: The server stores the received images and metadata and proceeds to the next analysis step. The input is the data package sent from the device, and the output is the stored image data and metadata.

[1293] Step 4: Analyze and select images

[1294] Server: The server uses an AI algorithm to analyze the received images. It selects the most suitable images based on criteria such as texture and color. The input is the stored image data and metadata, and the output is the selected image.

[1295] Step 5: GAN Processing

[1296] Server: The server generates a highly accurate final image using a generative anti-theft network (GAN) based on the filtered image. The input is the filtered image, and the output is the highly accurate final image.

[1297] Step 6: Generate and return test results

[1298] Server: Generates product inspection results based on the final image and sends the results back to the terminal. The input is the final image generated with high accuracy, and the output is the inspection result data.

[1299] Terminal: The terminal processes the inspection results received from the server and executes the next instructions on the production line. The input is the inspection result data from the server, and the output is the instructions for the production line.

[1300] Step 7: Compensation and Allocation

[1301] Server: The server checks the distributed network log and calculates rewards based on the provision of computing resources by each terminal. The input is the network log and computing resource information, and the output is the reward amount for each terminal.

[1302] User: The user who provides the terminal receives the reward in the form of electronic money. The input is the reward information from the server, and the output is a transfer to the user's electronic wallet.

[1303] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1304] As an embodiment of the present invention, we will explain a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. The system mainly operates in cooperation between a server, terminals, and users.

[1305] Server Features

[1306] 1. Network Construction

[1307] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[1308] 2. Task Distribution

[1309] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[1310] 3. Image collection and analysis

[1311] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[1312] 4. Emotional Engine Adjustment

[1313] The server uses an emotion engine to analyze the user's emotions in real time and adjusts image generation parameters and task distribution based on the analysis.

[1314] 5. GAN Processing

[1315] The server then uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs use two neural networks, a discriminator and a generator, to compete with each other to learn and generate more realistic images.

[1316] 6. Reward Allocation

[1317] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1318] Device Features

[1319] 1. Registering with the Network

[1320] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1321] 2. Receiving and processing tasks

[1322] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[1323] 3. Sending images

[1324] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[1325] User Roles

[1326] 1. Request for image generation

[1327] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[1328] 2. Providing emotions

[1329] Users use cameras and sensors to send facial and voice data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[1330] 3. Receiving rewards

[1331] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1332] Specific examples

[1333] Image generation request example

[1334] 1. Receiving a request

[1335] The server receives a request from the user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[1336] 2. Task Distribution and Processing

[1337] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[1338] 3. Image selection and GAN processing

[1339] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1340] 4. Emotional Engine Adjustment

[1341] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. Based on this data, the server adjusts the parameters required for the next image generation.

[1342] 5. Reward Allocation

[1343] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1344] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[1345] The processing flow will be explained below.

[1346] Step 1:

[1347] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[1348] Step 2:

[1349] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[1350] Step 3:

[1351] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[1352] Step 4:

[1353] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[1354] Step 5:

[1355] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[1356] Step 6:

[1357] Terminal A and terminal B process the received task and generate images. Terminal A generates an image with a sand texture, and terminal B generates an image with a mountain silhouette.

[1358] Step 7:

[1359] Once the generated images are complete, device A and device B send each generated image to the server, along with metadata such as the usage of computing resources.

[1360] Step 8:

[1361] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[1362] Step 9:

[1363] The server uses GAN to generate a highly accurate final image based on the selected images. In this process, two neural networks, a discriminator and a generator, compete with each other to learn and generate realistic images.

[1364] Step 10:

[1365] While the user evaluates the generated images, real-time emotional data is collected via a camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction.

[1366] Step 11:

[1367] The server adjusts the parameters required for the next image generation based on the data analyzed by the emotion engine. For example, if the user is not satisfied, it adjusts the color tone or details.

[1368] Step 12:

[1369] The server checks the distributed network logs and records which devices provided how much computing resources.

[1370] Step 13:

[1371] The server calculates rewards based on the amount of computing resources provided. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[1372] Step 14:

[1373] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[1374] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and has an adjustment function that reflects the user's emotional data in real time, ultimately achieving appropriate reward distribution.

[1375] Example 2

[1376] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1377] Conventional image generation systems lack the technology to rapidly generate highly accurate images or to generate images that reflect user emotions. They also lack the means to efficiently utilize distributed networks and distribute appropriate rewards based on computational resources. To address these issues, a system that integrates advanced image analysis and generation technology with the ability to reflect users' real-time emotions is needed.

[1378] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1379] In this invention, the server includes a means for constructing a distributed network, a means for supplying computing resources to each terminal, and a means for distributing image generation tasks. This enables highly accurate image generation and further enables real-time analysis of user emotions and their reflection in image generation. It also makes it possible to allocate appropriate rewards to each terminal based on the provision of computing resources.

[1380] A "distributed network" is a network structure in which multiple terminals work together to process data and communicate without relying on a centralized server.

[1381] "Computational resources" refers to the hardware and performance required for data processing and calculations, such as a computer system's CPU, GPU, memory, and storage.

[1382] An "image generation task" is a specific task or instruction for generating a new image using a particular algorithm or process.

[1383] "AI analysis and selection" means using artificial intelligence technology to analyze input data or information and select the best option based on specific criteria.

[1384] An "emotion engine" is a technology that analyzes emotions from a user's facial expressions and voice and adjusts the system's operation based on the results.

[1385] "GAN" stands for Generative Adversarial Network, a technology in which two neural networks, a generator and a discriminator, learn by competing with each other to generate highly accurate data (such as images).

[1386] "Electronic money" is digital or virtual currency that has value in electronic form and is traded over the Internet, rather than physical cash.

[1387] A "log" is data that represents a record of system actions or events, which can later be used for analysis or auditing.

[1388] This invention is a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. This system operates mainly through the cooperation of three elements: the server, the terminal, and the user. The roles and specific operations of each are described below.

[1389] Server Features

[1390] 1. Network Construction

[1391] The server creates a distributed network and sets up a public IP address for devices to join, allowing them to contact the server and register on the network.

[1392] 2. Task Distribution

[1393] The server receives an image generation request from a user, divides the request into small tasks, and distributes them to each terminal. At this time, it assigns the task to the most suitable terminal depending on the type and difficulty of the task.

[1394] 3. Image collection and analysis

[1395] The server collects the images generated by each device and uses AI algorithms to analyze and evaluate them, including texture and color, to select the best images.

[1396] 4. Emotional Engine Adjustment

[1397] The server uses an emotion engine to analyze the user's emotion data in real time and adjusts the image generation parameters based on the results, making it possible to generate images that are in line with the user's emotions.

[1398] 5. GAN Processing

[1399] The server uses a generative adversarial network (GAN) to generate a highly accurate final image based on the selected images. GANs are made up of two neural networks, a generator and a discriminator, that compete with each other to learn and generate more realistic images.

[1400] 6. Reward Allocation

[1401] The server checks the distributed network log and pays each terminal electronic money in return for the computing resources provided.

[1402] Device Features

[1403] 1. Registering with the Network

[1404] The device accesses the server and registers its computing resources (CPU / GPU performance, etc.) After registration, the device operates as part of the distributed network and is ready to receive tasks.

[1405] 2. Receiving and processing tasks

[1406] The device receives image generation tasks assigned by the server and performs the tasks, such as "generating the texture of sand" or "generating a night view of a city."

[1407] 3. Sending images

[1408] When the generated image is sent to the server, metadata such as the usage status of computing resources is also sent.

[1409] User Roles

[1410] 1. Request for image generation

[1411] A user requests the server to generate a specific image, specifically, by describing the request, such as a "mountain landscape" or a "city night view."

[1412] 2. Providing emotions

[1413] Users use cameras and sensors to send their emotional data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[1414] 3. Receiving rewards

[1415] Users who provide their devices receive rewards based on the tasks their devices complete, and the rewards are transferred to the users' electronic wallets as electronic money.

[1416] Specific examples

[1417] Image generation request example

[1418] 1. Receiving a request

[1419] The server receives a request from a user, such as "I want you to generate a desert landscape." The request is divided into smaller tasks, such as "generate the texture of sand" and "generate distant mountains."

[1420] 2. Task Distribution and Processing

[1421] The server assigns the task of "generating a sand texture" to terminal A and the task of "generating a silhouette of a distant mountain" to terminal B. Terminal A generates the sand texture, and terminal B generates the mountain silhouette, and they each send them to the server.

[1422] 3. Image selection and GAN processing

[1423] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1424] 4. Emotional Engine Adjustment

[1425] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. The server then adjusts the parameters required for the next image generation based on this data.

[1426] 5. Reward Allocation

[1427] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1428] Prompt Sentence Examples

[1429] "I want a desert landscape to be generated."

[1430] "I want to see the city's night view."

[1431] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[1432] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1433] Step 1:

[1434] Server: Network construction

[1435] The server sets up a distributed network by setting up a public IP address, which allows devices to join the network. The input requires the server's network configuration information, and the output is an initialized network. Specifically, the server edits the network configuration file, defines the public IP address and port number, and then listens on the network.

[1436] Step 2:

[1437] End user: Input of image generation request

[1438] A user requests a server to generate a specific image. The input is the user's request (e.g., "desert landscape"), and the output is the request being processed by the server. Specifically, the user enters the statement "I want a desert landscape generated" into a form in a web application and presses the submit button.

[1439] Step 3:

[1440] Server: Analyzes the request and distributes tasks

[1441] The server analyzes the received request and divides the task into smaller parts. The input is the user's request, and the output is the divided tasks. Specifically, the server divides the task into tasks such as "generate the texture of sand" and "generate the silhouette of a distant mountain," and distributes each task to the terminal.

[1442] Step 4:

[1443] Terminal: Receiving and processing tasks

[1444] The device receives tasks from the server and generates images. The input is the task content distributed by the server (e.g., "generate the texture of sand"), and the output is the generated image. Specifically, the device uses a deep learning model to generate the texture of sand and create its image data.

[1445] Step 5:

[1446] Device: Send image

[1447] The device sends the generated image to the server. The input is the generated image data, and the output is the image data sent to the server. Specifically, the device uploads the generated image data to the API endpoint specified by the server.

[1448] Step 6:

[1449] Server: Image collection and analysis

[1450] The server collects images received from the devices and analyzes them using an AI algorithm. The input is the image data sent from each device, and the output is the evaluated image data. Specifically, the server selects the most suitable images based on evaluation criteria such as the texture and color of each image.

[1451] Step 7:

[1452] Server: Coordination by emotion engine

[1453] The server analyzes the user's emotional data using an emotion engine and adjusts the image generation parameters. The input is the user's emotional data (for example, facial expression data and voice data), and the output is the adjusted parameters. Specifically, the server collects the user's emotional data, analyzes it using the emotion engine, and adjusts the parameters of the image generation algorithm based on the results.

[1454] Step 8:

[1455] Server: High-precision image generation using GAN processing

[1456] The server generates a high-precision image using a GAN model based on the selected images. The input is the evaluated image data, and the output is the final high-precision image. Specifically, the server inputs the selected image data into the generator, and has it compete with the discriminator to progress learning, ultimately generating a high-precision image.

[1457] Step 9:

[1458] Server: Reward Distribution

[1459] The server checks the distributed network log and calculates rewards based on the provision of computing resources. The input is the processing log data of each terminal, and the output is the reward amount for each terminal. Specifically, the server analyzes the processing log of each terminal and transfers the calculated electronic money as a reward to the electronic wallet of each terminal owner.

[1460] (Application example 2)

[1461] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1462] Conventional image generation systems require large amounts of computational resources to generate highly accurate images, and they have difficulty reflecting the specific emotional satisfaction desired by users in real time. Furthermore, there is a lack of a way to effectively utilize user feedback in the image generation process. As a result, images that do not fully meet the user's requirements are often generated, and satisfactory results cannot be achieved.

[1463] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and transmitting generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, means for collecting real-time emotional data from users, analyzing it, and reflecting it in the image generation process, and means for linking images generated via a terminal such as a smartphone with user emotional feedback. This makes it possible to generate high-precision images while also reflecting user emotional feedback in real time, thereby achieving image generation that greatly improves user satisfaction.

[1464] A "distributed network" is a network in which multiple devices are interconnected and can share and process data without centralized control.

[1465] "Computational resources" refers to the capabilities of hardware and software to perform computational processing, and specifically includes the performance of CPUs and GPUs.

[1466] "Image generation task" means a set of instructions for processing or computing required to generate a particular image.

[1467] A "server" is a central computer that processes and provides data in response to requests from multiple terminals within a network.

[1468] A "terminal" refers to a computer or device that is connected to a distributed network and provides computing resources.

[1469] "AI analysis and selection methods" refers to technology that uses artificial intelligence to evaluate collected images and select the most suitable ones.

[1470] "GAN" stands for generative adversarial network, a type of algorithm that generates realistic images by training two neural networks, a generator and a discriminator, to compete with each other.

[1471] "High-precision images" refer to detailed, high-quality images that closely resemble real-world scenes and objects.

[1472] "Electronic money" means currency provided in digital form, an asset that is exchanged electronically.

[1473] "Real-time emotional data" refers to information that collects the user's emotional state in real time and processes that data immediately.

[1474] "Emotional feedback" refers to the evaluation and response of a system based on emotional expressions and data provided by the user.

[1475] This invention is a system that uses a distributed network to generate highly accurate images, and also has the function of recognizing and reflecting user emotions in real time. The following is a specific embodiment of this invention.

[1476] Server Features

[1477] The server has the following main functions:

[1478] 1. Networking:

[1479] The server creates a distributed network and configures it so that each terminal can provide computing resources. It uses public IP addresses to provide an environment where terminals can easily connect.

[1480] 2. Task distribution:

[1481] The system divides image generation requests received from users into small tasks and distributes them to each device. For example, it gives detailed instructions such as "generate the texture of sand" or "generate the silhouette of a distant mountain" to each device.

[1482] 3. Image Collection and Analysis:

[1483] Images generated by each device are collected and analyzed and evaluated using AI algorithms, and the best images are selected based on evaluation criteria such as texture, color, and composition.

[1484] 4. Emotional engine regulation:

[1485] The system analyzes the user's emotional data in real time and reflects it at each stage of the generation process. This is achieved by analyzing the emotional information provided by the user through a camera or microphone.

[1486] 5. GAN Processing:

[1487] To generate the final high-precision image, a generative adversarial network (GAN) is used based on the selected images. GAN includes two neural networks: a generator and a discriminator.

[1488] 6. Reward Allocation:

[1489] The system checks the distributed network logs, calculates rewards based on the computing resources provided, and pays each device electronic money.

[1490] Device Features

[1491] The terminal has the following features:

[1492] 1. Registering with the Network:

[1493] They connect to a server, register with the decentralized network, provide computing resources, and receive tokens in return.

[1494] 2. Receiving and processing tasks:

[1495] It receives and processes image generation tasks from the server, and includes metadata such as the usage of computing resources when sending the generated images to the server.

[1496] User Roles

[1497] Users have the following roles:

[1498] 1. Image generation request:

[1499] Request the server to generate a specific image, specifying specific content such as a "mountain landscape" or a "city night view."

[1500] 2. Providing Emotion Data:

[1501] Cameras and sensors are used to provide facial and voice data in real time, which is analyzed by the emotion engine and reflected in image generation.

[1502] Specific examples

[1503] A user opens the app and requests that an image of a "summer beach sunset" be generated. The app collects the user's emotional data in real time via the camera and microphone. When the user expresses emotions such as "happy" or "moved" while looking at the image, that data is sent to the server and reflected in the image generation. Finally, an image that satisfies the user is generated and displayed on the smartphone.

[1504] An example prompt for a generative AI model is:

[1505] "Generate a summer seaside scene. I'd like the image to have a tranquil, relaxing atmosphere with the sound of the waves. Additionally, I'd like the sunset to be reflected on the water to enhance its beauty."

[1506] The above is an embodiment of the present invention. This system not only generates highly accurate images but also reflects real-time emotional feedback from the user, thereby improving user satisfaction.

[1507] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1508] Step 1:

[1509] Request Receipt

[1510] The server receives an image generation request from a user.

[1511] (Input) The user's request (e.g., "Summer sunset at the beach").

[1512] (Output) Initial settings for the image generation task.

[1513] (Specific operation) A user requests the generation of a specific image through the app, and the server receives the request, which may include a prompt for the generating AI model.

[1514] Step 2:

[1515] Task division

[1516] The server divides the image generation request into multiple subtasks.

[1517] (Input) The received image generation request.

[1518] (Output) A set of divided image generation tasks.

[1519] (Specific operation) The server breaks down the data into specific parts such as "generating the texture of sand" or "adjusting the color of the sunset" and converts each part into a processable format.

[1520] Step 3:

[1521] Task Distribution

[1522] The server distributes the divided tasks to the appropriate terminals.

[1523] (Input) A set of divided image generation tasks.

[1524] (Output) Tasks sent to each device.

[1525] (Specific operation) The server assigns tasks and sends instructions based on each terminal's available computing resources and past performance.

[1526] Step 4:

[1527] Receiving and processing tasks

[1528] The terminal processes the received task and sends the results to the server.

[1529] (Input) Image generation task sent from the server.

[1530] (Output) The generated subimage and metadata.

[1531] (Specific operation) Each device executes the received task, specifically generating the texture and color of a specific image, and sends this to the server along with metadata including the usage status of computing resources.

[1532] Step 5:

[1533] Image collection and analysis

[1534] The server collects images sent from each device and analyzes them using AI algorithms.

[1535] (Input) Partial images and metadata sent from each device.

[1536] (Output) The analyzed image data.

[1537] (Specific operation) The server evaluates the texture, color, composition, etc. of each image and selects the most suitable image using an AI algorithm.

[1538] Step 6:

[1539] Emotion data collection and analysis

[1540] The server collects and analyzes the user's real-time emotional data.

[1541] (Input) Emotion data from the user via camera and microphone.

[1542] (Output) Parsed emotion information.

[1543] (Specific operation) While the user is checking the generated image, facial expression and voice data is collected via the smartphone's camera and microphone, and the server analyzes this data using the emotion engine.

[1544] Step 7:

[1545] Reflecting and adjusting emotional data

[1546] The server adjusts image generation parameters based on the analyzed emotional information.

[1547] (Input) Analyzed emotion information.

[1548] (Output) Adjusted image generation parameters.

[1549] (Specific operation) The server adjusts the parameters of the generation process according to the user's positive or negative emotions. This adjustment ensures that the next task executed reflects the emotional data.

[1550] Step 8:

[1551] Final GAN ​​processing

[1552] The server generates the final image using GAN based on the selected images.

[1553] (Input) Selected partial images and adjusted parameters.

[1554] (Output) High-precision final image.

[1555] (Specific operation) Using a GAN (generative adversarial network) with a generator and discriminator, the final result is a highly accurate and realistic image.

[1556] Step 9:

[1557] Final image courtesy

[1558] The server provides the generated final image to the user.

[1559] (Input) High-precision final image.

[1560] (Output) The image that is displayed on the user device.

[1561] (Specific operation) The server sends the generated final image to the user's device, and the app displays it.

[1562] Step 10:

[1563] Reward Allocation

[1564] The server distributes rewards to the terminals that participate in the task processing.

[1565] (Input) Distributed network log and contribution data of each device.

[1566] (Output) Reward payment to each device.

[1567] (Specific operation) The server analyzes the logs of the distributed network, calculates rewards based on the degree of contribution, and pays electronic money to each terminal.

[1568] This concludes the explanation of the specific processing steps. This system enables highly accurate image generation and reflects the user's emotional feedback in real time.

[1569] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1570] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1571] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1572] [Fourth embodiment]

[1573] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1574] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1575] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1576] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1577] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1578] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1579] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1580] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1581] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1582] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1583] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1584] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1585] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1586] As an embodiment of the present invention, a system for generating highly accurate images using a distributed network and paying rewards to terminals that contribute to the image generation will be described below. The system operates mainly in cooperation between a server, terminals, and users.

[1587] Server Features

[1588] 1. Network Construction

[1589] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[1590] 2. Task Distribution

[1591] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[1592] 3. Image collection and analysis

[1593] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[1594] 4. GAN Processing

[1595] The server uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs are made up of two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[1596] 5. Reward Allocation

[1597] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1598] Device Features

[1599] 1. Registering with the Network

[1600] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1601] 2. Receiving and processing tasks

[1602] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[1603] 3. Sending images

[1604] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[1605] User Roles

[1606] 1. Request for image generation

[1607] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[1608] 2. Receiving Rewards

[1609] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1610] Specific examples

[1611] Image generation request example

[1612] 1. Receiving a request

[1613] The server receives a request from a user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[1614] 2. Task Distribution and Processing

[1615] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[1616] 3. Image selection and GAN processing

[1617] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1618] 4. Reward Allocation

[1619] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1620] The above is a detailed description of the embodiment of the present invention. This system realizes efficient and highly accurate image generation and appropriate reward distribution using a distributed network.

[1621] The processing flow will be explained below.

[1622] Step 1:

[1623] The server creates a distributed network and configures it so that nodes (devices) can join, which includes setting up public IP addresses so that devices can connect to the network.

[1624] Step 2:

[1625] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[1626] Step 3:

[1627] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[1628] Step 4:

[1629] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[1630] Step 5:

[1631] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[1632] Step 6:

[1633] The devices process the received tasks and generate images. Device A generates an image with a sandy texture, while device B generates an image with a mountain silhouette.

[1634] Step 7:

[1635] The device sends the generated images to the server, and metadata such as computational resource usage is included with each image.

[1636] Step 8:

[1637] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[1638] Step 9:

[1639] The server then uses GAN to generate a highly accurate final image based on the selected images. GAN uses two neural networks, a discriminator and a generator, that compete with each other to learn and generate more realistic images.

[1640] Step 10:

[1641] The server checks the distributed network logs and records which devices provided how much computing resources.

[1642] Step 11:

[1643] The server calculates rewards based on the provision of computing resources. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[1644] Step 12:

[1645] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[1646] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and achieves appropriate reward distribution.

[1647] Example 1

[1648] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1649] This invention relates to a system that efficiently realizes high-precision image generation and fairly distributes rewards to terminals that provide computational resources that contribute to image generation. Conventional systems mainly use a centralized approach, which has led to problems such as the concentration of computational resources and unfair distribution of rewards. Furthermore, high-precision image generation requires a large amount of computational resources, which calls for a distributed computing environment.

[1650] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1651] In this invention, the server includes a means for constructing a distributed network, a means for providing computational resources to each terminal, and a means for receiving and decomposing image generation requests. This enables efficient high-precision image generation and fair distribution of rewards to terminals that provide computational resources. Specifically, the server assigns image generation tasks to each terminal, evaluates the generated images, and generates a high-precision final image. The server also checks the distributed network log, calculates rewards based on the provision of computational resources, and pays them in electronic money, thereby achieving overall system efficiency and fairness.

[1652] A "distributed network" is a network in which terminals with multiple computing resources are connected on an equal basis and perform data processing and communication without central control.

[1653] "Computational resources" is a general term for hardware resources used for data processing and calculations, such as a terminal's CPU, GPU, and memory.

[1654] An "image generation request" is a request or prompt that a user sends to a server to generate a particular image.

[1655] An "image generation task" is a small unit of work required to realize an image generation request, and is distributed to each terminal.

[1656] "Generative network" refers to a generative adversarial network (GAN), an algorithm in which two neural networks, a discriminator and a generator, compete to generate data.

[1657] "Task allocation" refers to the operations and processes for appropriately distributing image generation tasks to each terminal.

[1658] "Reward" refers to the compensation paid to the terminal that contributed to the image generation, and is usually provided in the form of electronic money.

[1659] "Electronic money" means monetary value provided in digital form and used for online transactions and payment of rewards.

[1660] The present invention provides a system that uses a distributed network to achieve high-precision image generation and fairly distributes rewards to terminals that contribute to image generation. Specific embodiments for implementing this system are described below.

[1661] Hardware and Software Use

[1662] In this system, the server uses dedicated software to build and manage the distributed network. Specifically, distributed network software such as Apache Kafka and Ethereum is used. The server has computational resources such as CPU, GPU, and memory, and functions as a system to manage these. The terminals are computers or smart devices that provide the computational resources. The terminals use local resources (CPU / GPU) to process image generation tasks.

[1663] Processing flow

[1664] 1. Network Construction

[1665] The server builds a distributed network and sets up a public IP address to provide an environment in which nodes (terminals) can participate. The server also authenticates nodes and allows them to connect to the network safely.

[1666] 2. Receiving an image generation request

[1667] The user sends a request to the server to generate an image. For example, the user can send a prompt such as "I want you to generate a sunset ocean image."

[1668] 3. Task Decomposition and Allocation

[1669] The server receives an image generation request and divides it into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset." The server distributes these tasks according to the resource availability of each device.

[1670] 4. Task processing and image generation

[1671] The device processes tasks assigned by the server. For example, device A receives the task of "generating sky color" and uses its GPU to generate realistic sky colors. The generated partial images are stored in local storage.

[1672] 5. Image Collection and Evaluation

[1673] The server collects the partial images sent from each device, evaluates them using AI algorithms (such as ResNet or Inception), and selects the best images based on criteria such as texture and color.

[1674] 6. GAN processing and final image generation

[1675] The server generates a highly accurate final image based on the selected images using a Generative Adversarial Network (GAN), which consists of two neural networks, a discriminator and a generator, that learn by competing with each other.

[1676] 7. Reward Allocation

[1677] The server checks the distributed network log and collates which terminal provided how much computing resources. Based on this, it pays the user who provided the terminal a reward in the form of electronic money. The reward is transferred to the user's electronic wallet.

[1678] Specific examples

[1679] Prompt Sentence Examples

[1680] "I want to generate a sunset ocean"

[1681] In this way, our system can efficiently generate highly accurate images and distribute rewards fairly based on the computing resources provided by the terminals.

[1682] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1683] Processing Steps

[1684] Step 1: Building the network

[1685] 1. Specific actions

[1686] To build a distributed network, the server must install and configure dedicated software (e.g., Apache Kafka or Ethereum), set up a public IP address, and create an environment where connections can be made.

[1687] 2. Input

[1688] Necessary network configuration information, software configuration files

[1689] 3. Output

[1690] A fully distributed network, with devices ready to connect

[1691] 4. Detailed explanation

[1692] The server sets up a terminal authentication system to ensure that newly joining terminals can securely connect to the network. To maintain the health of the network, the server monitors the node status in real time.

[1693] Step 2: Receiving an image generation request

[1694] 1. Specific actions

[1695] The user sends a request to the server to generate an image. For example, the user inputs a prompt such as "I want to generate a sunset ocean image."

[1696] 2. Input

[1697] User prompt (e.g. "I want to generate a sunset ocean")

[1698] 3. Output

[1699] Image generation requests recorded in the server's internal database

[1700] 4. Detailed explanation

[1701] The server analyzes the prompt received from the user and records its contents in an internal database. Based on this information, task decomposition is performed in the next processing step.

[1702] Step 3: Break down and assign tasks

[1703] 1. Specific actions

[1704] The server divides the image generation request into multiple small tasks, such as "generate the color of the sky," "generate the texture of waves," and "generate the reflection of the sunset," and assigns each task to a device.

[1705] 2. Input

[1706] Image generation request from user, device resource information

[1707] 3. Output

[1708] Individual image generation tasks distributed to devices

[1709] 4. Detailed explanation

[1710] The server allocates tasks efficiently, taking into account the resources of each device (CPU, GPU, memory usage), and notifies the device of the allocated tasks.

[1711] Step 4: Receiving and Processing Tasks

[1712] 1. Specific actions

[1713] The device receives tasks assigned by the server and adds them to a local queue. The device executes the tasks using the GPU or CPU to generate images.

[1714] 2. Input

[1715] Image generation tasks assigned by the server

[1716] 3. Output

[1717] Generated partial images, metadata (resources used, processing time, etc.)

[1718] 4. Detailed explanation

[1719] The device uses a local AI model (e.g., CNN or Transformer) based on the task content to generate the specified image elements, and the generated data is stored in local storage.

[1720] Step 5: Collect and evaluate images

[1721] 1. Specific actions

[1722] The server receives the partial images generated by each device and evaluates them using an AI algorithm (e.g., ResNet, Inception).

[1723] 2. Input

[1724] Partial images and metadata sent from the device

[1725] 3. Output

[1726] Evaluation results and selected optimal images

[1727] 4. Detailed explanation

[1728] The server analyzes the texture, color, sharpness, etc. of the collected images and selects the best images based on the evaluation criteria. This selection is used for the next GAN processing.

[1729] Step 6: GAN processing and final image generation

[1730] 1. Specific actions

[1731] The server uses the selected images as input data and generates a highly accurate final image using a Generative Adversarial Network (GAN).

[1732] 2. Input

[1733] Evaluated and selected optimal images

[1734] 3. Output

[1735] High-precision final image

[1736] 4. Detailed explanation

[1737] GAN involves two neural networks, a discriminator and a generator, competing to learn and ultimately generating highly accurate, realistic images, which are then stored in server storage.

[1738] Step 7: Reward Allocation

[1739] 1. Specific actions

[1740] The server checks the distributed network log and calculates the contribution of each device, and based on this, it pays electronic money to the users who provided their devices.

[1741] 2. Input

[1742] Distributed network log, contribution data of each device

[1743] 3. Output

[1744] Rewards sent to the user's electronic wallet

[1745] 4. Detailed explanation

[1746] The server analyzes the usage status of computing resources and processing results of each device, and calculates rewards based on that. The calculated rewards are transferred to the user's electronic wallet as electronic money, realizing fair reward distribution according to the degree of contribution.

[1747] (Application example 1)

[1748] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1749] The problem that the present invention aims to solve is to provide a system that can perform product quality inspections efficiently and with high accuracy in the manufacturing industry. Conventional product inspection methods have limitations in inspection efficiency and accuracy, and require a great deal of effort and cost. Furthermore, conventional methods often involve the subjective judgment of the inspector, making consistent quality control difficult.

[1750] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1751] In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and sending the generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, and means for using the generated images for product inspection and returning the inspection results to the robot. This enables highly accurate and efficient product quality inspection in the manufacturing industry.

[1752] A "distributed network" is a network in which multiple terminals provide computing resources in a distributed manner and work together to process a single task.

[1753] "Computational resources" refer to computer resources such as CPU and GPU performance, memory capacity, and storage capacity provided by a terminal.

[1754] An "image generation task" is a unit of work that includes specific calculations and processing to generate an image.

[1755] "AI" is an abbreviation for artificial intelligence, a technology that uses algorithms such as machine learning and neural networks to analyze data and behave intelligently.

[1756] "Selection" is the process of selecting the best images from the generated images based on evaluation criteria such as texture and color.

[1757] "GAN" is an abbreviation for Generative Adversarial Network, an algorithm in which two neural networks (a generator and a discriminator) compete with each other to learn and generate highly accurate data.

[1758] "Remuneration" is the compensation paid to a terminal that provides computing resources within the system, and is often in the form of electronic money.

[1759] "Product inspection" is the process of checking the quality and condition of manufactured products to ensure they meet standards.

[1760] A "robot" is a device that operates mechanically to perform designated tasks. It is often used in automated manufacturing processes.

[1761] MODE FOR CARRYING OUT THE INVENTION

[1762] As an embodiment of the present invention, a system for performing high-precision product inspection in a factory using a distributed network will be described. This system operates mainly in cooperation between a server, a terminal (robot), and a user.

[1763] Server Features

[1764] 1. Networking:

[1765] The server creates a distributed network and allows nodes (terminals) to participate. The server sets up a public IP address so that participating terminals can connect to the network.

[1766] 2. Task distribution:

[1767] The server receives product inspection requests from users and breaks them down into smaller tasks, such as inspecting specific parts or checking specific details. It then distributes these tasks to each terminal and manages them appropriately.

[1768] 3. Image Collection and Analysis:

[1769] The server collects product images acquired from each device, analyzes and evaluates the collected images using AI algorithms, and selects the most suitable images based on criteria such as the presence or absence of defects.

[1770] 4. GAN Processing:

[1771] The server uses GAN (Generic Annotation) to generate a highly accurate final image based on the selected images. GAN consists of two neural networks, a discriminator and a generator, that compete with each other to learn and provide more accurate inspection results.

[1772] 5. Reward Allocation:

[1773] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1774] 6. Return of test results:

[1775] The server compiles the test results and sends the final test results back to the robot.

[1776] Terminal (robot) functions

[1777] 1. Registering with the Network:

[1778] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1779] 2. Receiving and processing tasks:

[1780] The terminal receives and processes image generation and inspection tasks assigned to it by the server, performing specific tasks such as capturing images to inspect a specific part or evaluating the generated images to identify specific details.

[1781] 3. Sending images:

[1782] The device captures an image of the product and sends it to a server, along with metadata such as computing resource usage.

[1783] User Roles

[1784] 1. Product Inspection Request:

[1785] The user requests a specific product inspection from the server, specifying, for example, "Please inspect the surface of the part for scratches or dirt" or "Please check the product for color unevenness or shape abnormalities."

[1786] 2. Receiving rewards:

[1787] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1788] Hardware and software used

[1789] Hardware:

[1790] A robot installed on a factory production line (equipped with an image capture camera)

[1791] Modem / router for network connection

[1792] software:

[1793] Implemented server and robot control programs using Python.

[1794] Use the Requests library to process HTTP requests.

[1795] Use a local server or a cloud server (e.g. AWS, Google Cloud) on the server side.

[1796] Examples:

[1797] For example, one product inspection request might be "surface scratch inspection." The server splits this request into two tasks: "capturing an image of a specific part" and "analyzing the captured image using AI." The captured image is sent to the server and analyzed by an AI algorithm. The analysis results are used to determine whether there are any surface scratches using a deep learning model, and the results are sent back to the robot. The robot then decides on the next action to take on the production line.

[1798] Example prompt: "Inspect the product surface for scratches or stains."

[1799] This system enables factories to automate product inspection with high precision and efficiency, dramatically improving quality control.

[1800] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1801] Step 1: Register with the network

[1802] Server: The server sets a public IP address and builds a distributed network. It prepares the network environment so that devices can join. The input is the public IP address and network setting information, and the output is a network environment that devices can connect to.

[1803] Device: The device accesses the server to register on the network. As registration information, it sends its device ID and performance (CPU / GPU specs). The input is the device ID and performance information, and the output is a network participation token issued by the server.

[1804] Step 2: Receiving and processing the image generation task

[1805] Server: The server receives product inspection requests from users and divides them into image generation tasks. Tasks include inspecting specific parts and checking details. The input is the inspection request from the user, and the output is the image generation tasks assigned to each terminal.

[1806] Terminal: The terminal processes tasks received from the server. For example, it takes an image of a specific part of a product and sends the captured data to the server. The input is the image generation task from the server, and the output is the captured product image.

[1807] Step 3: Submit your images and add metadata

[1808] Terminal: When generating product images, metadata such as the usage of computing resources is added to the images when they are sent to the server. The input is the captured product image and metadata, and the output is a data package sent to the server.

[1809] Server: The server stores the received images and metadata and proceeds to the next analysis step. The input is the data package sent from the device, and the output is the stored image data and metadata.

[1810] Step 4: Analyze and select images

[1811] Server: The server uses an AI algorithm to analyze the received images. It selects the most suitable images based on criteria such as texture and color. The input is the stored image data and metadata, and the output is the selected image.

[1812] Step 5: GAN Processing

[1813] Server: The server generates a highly accurate final image using a generative anti-theft network (GAN) based on the filtered image. The input is the filtered image, and the output is the highly accurate final image.

[1814] Step 6: Generate and return test results

[1815] Server: Generates product inspection results based on the final image and sends the results back to the terminal. The input is the final image generated with high accuracy, and the output is the inspection result data.

[1816] Terminal: The terminal processes the inspection results received from the server and executes the next instructions on the production line. The input is the inspection result data from the server, and the output is the instructions for the production line.

[1817] Step 7: Compensation and Allocation

[1818] Server: The server checks the distributed network log and calculates rewards based on the provision of computing resources by each terminal. The input is the network log and computing resource information, and the output is the reward amount for each terminal.

[1819] User: The user who provides the terminal receives the reward in the form of electronic money. The input is the reward information from the server, and the output is a transfer to the user's electronic wallet.

[1820] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1821] As an embodiment of the present invention, we will explain a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. The system mainly operates in cooperation between a server, terminals, and users.

[1822] Server Features

[1823] 1. Network Construction

[1824] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[1825] 2. Task Distribution

[1826] The server receives an image generation request from a user, divides it into small tasks, and distributes these tasks to each terminal and manages them appropriately.

[1827] 3. Image collection and analysis

[1828] The server collects images generated by each device, analyzes and evaluates them using an AI algorithm, and selects the most suitable images based on criteria such as texture and color.

[1829] 4. Emotional Engine Adjustment

[1830] The server uses an emotion engine to analyze the user's emotions in real time and adjusts image generation parameters and task distribution based on the analysis.

[1831] 5. GAN Processing

[1832] The server then uses the selected images to generate a highly accurate final image using a generative adversarial network (GAN). GANs use two neural networks, a discriminator and a generator, to compete with each other to learn and generate more realistic images.

[1833] 6. Reward Allocation

[1834] The server checks the distributed network log and records which terminals provided how much computing resources. Based on this, it pays each terminal a reward in the form of electronic money.

[1835] Device Features

[1836] 1. Registering with the Network

[1837] The device accesses the server and registers with the decentralized network. After registration, the device provides computing resources (CPU / GPU performance) and receives tokens.

[1838] 2. Receiving and processing tasks

[1839] The device receives image generation tasks assigned by the server and processes them, for example, executing specific tasks such as "generating a specific part of a landscape" or "generating a specific texture."

[1840] 3. Sending images

[1841] The generated image is sent to the server, along with metadata such as the usage status of computing resources.

[1842] User Roles

[1843] 1. Request for image generation

[1844] A user requests the server to generate a specific image, such as a "mountain landscape" or a "night view of a city," according to the user's requirements.

[1845] 2. Providing emotions

[1846] Users use cameras and sensors to send facial and voice data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[1847] 3. Receiving rewards

[1848] The user who provides the terminal receives a reward based on the tasks processed by the terminal, and this reward is transferred to the user's electronic wallet in the form of electronic money.

[1849] Specific examples

[1850] Image generation request example

[1851] 1. Receiving a request

[1852] The server receives a request from the user, such as "generate a desert landscape." The server breaks this request down into smaller tasks, such as "generate the texture of sand" and "generate the silhouette of a distant mountain."

[1853] 2. Task Distribution and Processing

[1854] The task "Generate the texture of sand" is distributed to terminal A, and the task "Generate the silhouette of a distant mountain" is distributed to terminal B. Terminal A generates the texture of sand, and terminal B generates the silhouette of a mountain, and they each send them to the server.

[1855] 3. Image selection and GAN processing

[1856] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1857] 4. Emotional Engine Adjustment

[1858] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. Based on this data, the server adjusts the parameters required for the next image generation.

[1859] 5. Reward Allocation

[1860] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1861] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[1862] The processing flow will be explained below.

[1863] Step 1:

[1864] The server creates a distributed network and configures it so that nodes (terminals) can participate. In this process, it sets public IP addresses so that terminals can connect to the network.

[1865] Step 2:

[1866] The device accesses the server and registers with the distributed network. During registration, the device provides information about its computing resources (CPU / GPU performance) and receives an authentication token from the server.

[1867] Step 3:

[1868] A user inputs a request for image generation to the server. For example, the user inputs a request for "generating a desert landscape."

[1869] Step 4:

[1870] The server receives an image generation request and breaks it down into smaller tasks, such as "generating the texture of sand" or "generating the silhouette of a distant mountain."

[1871] Step 5:

[1872] The server distributes each divided task to devices registered on the network. For example, "generating the texture of sand" is assigned to device A, and "generating the silhouette of a distant mountain" is assigned to device B.

[1873] Step 6:

[1874] Terminal A and terminal B process the received task and generate images. Terminal A generates an image with a sand texture, and terminal B generates an image with a mountain silhouette.

[1875] Step 7:

[1876] Once the generated images are complete, device A and device B send each generated image to the server, along with metadata such as the usage of computing resources.

[1877] Step 8:

[1878] The server analyzes the received images using AI algorithms and selects the best images based on criteria such as texture and color.

[1879] Step 9:

[1880] The server uses GAN to generate a highly accurate final image based on the selected images. In this process, two neural networks, a discriminator and a generator, compete with each other to learn and generate realistic images.

[1881] Step 10:

[1882] While the user evaluates the generated images, real-time emotional data is collected via a camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction.

[1883] Step 11:

[1884] The server adjusts the parameters required for the next image generation based on the data analyzed by the emotion engine. For example, if the user is not satisfied, it adjusts the color tone or details.

[1885] Step 12:

[1886] The server checks the distributed network logs and records which devices provided how much computing resources.

[1887] Step 13:

[1888] The server calculates rewards based on the amount of computing resources provided. For example, a device that uses more resources to generate a more accurate image will receive a higher reward.

[1889] Step 14:

[1890] The server pays the reward to the terminal (user) in electronic money, which is sent to the user's electronic wallet.

[1891] As a result, this system utilizes a distributed network to generate images efficiently and with high accuracy, and has an adjustment function that reflects the user's emotional data in real time, ultimately achieving appropriate reward distribution.

[1892] Example 2

[1893] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1894] Conventional image generation systems lack the technology to rapidly generate highly accurate images or to generate images that reflect user emotions. They also lack the means to efficiently utilize distributed networks and distribute appropriate rewards based on computational resources. To address these issues, a system that integrates advanced image analysis and generation technology with the ability to reflect users' real-time emotions is needed.

[1895] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1896] In this invention, the server includes a means for constructing a distributed network, a means for supplying computing resources to each terminal, and a means for distributing image generation tasks. This enables highly accurate image generation and further enables real-time analysis of user emotions and their reflection in image generation. It also makes it possible to allocate appropriate rewards to each terminal based on the provision of computing resources.

[1897] A "distributed network" is a network structure in which multiple terminals work together to process data and communicate without relying on a centralized server.

[1898] "Computational resources" refers to the hardware and performance required for data processing and calculations, such as a computer system's CPU, GPU, memory, and storage.

[1899] An "image generation task" is a specific task or instruction for generating a new image using a particular algorithm or process.

[1900] "AI analysis and selection" means using artificial intelligence technology to analyze input data or information and select the best option based on specific criteria.

[1901] An "emotion engine" is a technology that analyzes emotions from a user's facial expressions and voice and adjusts the system's operation based on the results.

[1902] "GAN" stands for Generative Adversarial Network, a technology in which two neural networks, a generator and a discriminator, learn by competing with each other to generate highly accurate data (such as images).

[1903] "Electronic money" is digital or virtual currency that has value in electronic form and is traded over the Internet, rather than physical cash.

[1904] A "log" is data that represents a record of system actions or events, which can later be used for analysis or auditing.

[1905] This invention is a system that uses a distributed network to generate highly accurate images and further adjusts the generation process by combining an emotion engine that recognizes the user's emotions. This system operates mainly through the cooperation of three elements: the server, the terminal, and the user. The roles and specific operations of each are described below.

[1906] Server Features

[1907] 1. Network Construction

[1908] The server creates a distributed network and sets up a public IP address for devices to join, allowing them to contact the server and register on the network.

[1909] 2. Task Distribution

[1910] The server receives an image generation request from a user, divides the request into small tasks, and distributes them to each terminal. At this time, it assigns the task to the most suitable terminal depending on the type and difficulty of the task.

[1911] 3. Image collection and analysis

[1912] The server collects the images generated by each device and uses AI algorithms to analyze and evaluate them, including texture and color, to select the best images.

[1913] 4. Emotional Engine Adjustment

[1914] The server uses an emotion engine to analyze the user's emotion data in real time and adjusts the image generation parameters based on the results, making it possible to generate images that are in line with the user's emotions.

[1915] 5. GAN Processing

[1916] The server uses a generative adversarial network (GAN) to generate a highly accurate final image based on the selected images. GANs are made up of two neural networks, a generator and a discriminator, that compete with each other to learn and generate more realistic images.

[1917] 6. Reward Allocation

[1918] The server checks the distributed network log and pays each terminal electronic money in return for the computing resources provided.

[1919] Device Features

[1920] 1. Registering with the Network

[1921] The device accesses the server and registers its computing resources (CPU / GPU performance, etc.) After registration, the device operates as part of the distributed network and is ready to receive tasks.

[1922] 2. Receiving and processing tasks

[1923] The device receives image generation tasks assigned by the server and performs the tasks, such as "generating the texture of sand" or "generating a night view of a city."

[1924] 3. Sending images

[1925] When the generated image is sent to the server, metadata such as the usage status of computing resources is also sent.

[1926] User Roles

[1927] 1. Request for image generation

[1928] A user requests the server to generate a specific image, specifically, by describing the request, such as a "mountain landscape" or a "city night view."

[1929] 2. Providing emotions

[1930] Users use cameras and sensors to send their emotional data to the server, which is then analyzed by the emotion engine and reflected in real-time image generation.

[1931] 3. Receiving rewards

[1932] Users who provide their devices receive rewards based on the tasks their devices complete, and the rewards are transferred to the users' electronic wallets as electronic money.

[1933] Specific examples

[1934] Image generation request example

[1935] 1. Receiving a request

[1936] The server receives a request from a user, such as "I want you to generate a desert landscape." The request is divided into smaller tasks, such as "generate the texture of sand" and "generate distant mountains."

[1937] 2. Task Distribution and Processing

[1938] The server assigns the task of "generating a sand texture" to terminal A and the task of "generating a silhouette of a distant mountain" to terminal B. Terminal A generates the sand texture, and terminal B generates the mountain silhouette, and they each send them to the server.

[1939] 3. Image selection and GAN processing

[1940] The server evaluates the received images using an AI algorithm and selects the best image, then uses GAN to generate a highly accurate image based on the selected image.

[1941] 4. Emotional Engine Adjustment

[1942] While the user evaluates the generated image, real-time emotional data is collected via the camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to detect dissatisfaction or satisfaction. The server then adjusts the parameters required for the next image generation based on this data.

[1943] 5. Reward Allocation

[1944] The server calculates the reward based on the task processing status of each terminal and pays the reward to terminal A and terminal B in electronic money.

[1945] Prompt Sentence Examples

[1946] "I want a desert landscape to be generated."

[1947] "I want to see the city's night view."

[1948] This concludes the detailed description of the embodiment of the present invention. This system combines efficient and highly accurate image generation using a distributed network with an adjustment function that reflects the user's emotions in real time, ultimately achieving appropriate reward distribution.

[1949] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1950] Step 1:

[1951] Server: Network construction

[1952] The server sets up a distributed network by setting up a public IP address, which allows devices to join the network. The input requires the server's network configuration information, and the output is an initialized network. Specifically, the server edits the network configuration file, defines the public IP address and port number, and then listens on the network.

[1953] Step 2:

[1954] End user: Input of image generation request

[1955] A user requests a server to generate a specific image. The input is the user's request (e.g., "desert landscape"), and the output is the request being processed by the server. Specifically, the user enters the statement "I want a desert landscape generated" into a form in a web application and presses the submit button.

[1956] Step 3:

[1957] Server: Analyzes the request and distributes tasks

[1958] The server analyzes the received request and divides the task into smaller parts. The input is the user's request, and the output is the divided tasks. Specifically, the server divides the task into tasks such as "generate the texture of sand" and "generate the silhouette of a distant mountain," and distributes each task to the terminal.

[1959] Step 4:

[1960] Terminal: Receiving and processing tasks

[1961] The device receives tasks from the server and generates images. The input is the task content distributed by the server (e.g., "generate the texture of sand"), and the output is the generated image. Specifically, the device uses a deep learning model to generate the texture of sand and create its image data.

[1962] Step 5:

[1963] Device: Send image

[1964] The device sends the generated image to the server. The input is the generated image data, and the output is the image data sent to the server. Specifically, the device uploads the generated image data to the API endpoint specified by the server.

[1965] Step 6:

[1966] Server: Image collection and analysis

[1967] The server collects images received from the devices and analyzes them using an AI algorithm. The input is the image data sent from each device, and the output is the evaluated image data. Specifically, the server selects the most suitable images based on evaluation criteria such as the texture and color of each image.

[1968] Step 7:

[1969] Server: Coordination by emotion engine

[1970] The server analyzes the user's emotional data using an emotion engine and adjusts the image generation parameters. The input is the user's emotional data (for example, facial expression data and voice data), and the output is the adjusted parameters. Specifically, the server collects the user's emotional data, analyzes it using the emotion engine, and adjusts the parameters of the image generation algorithm based on the results.

[1971] Step 8:

[1972] Server: High-precision image generation using GAN processing

[1973] The server generates a high-precision image using a GAN model based on the selected images. The input is the evaluated image data, and the output is the final high-precision image. Specifically, the server inputs the selected image data into the generator, and has it compete with the discriminator to progress learning, ultimately generating a high-precision image.

[1974] Step 9:

[1975] Server: Reward Distribution

[1976] The server checks the distributed network log and calculates rewards based on the provision of computing resources. The input is the processing log data of each terminal, and the output is the reward amount for each terminal. Specifically, the server analyzes the processing log of each terminal and transfers the calculated electronic money as a reward to the electronic wallet of each terminal owner.

[1977] (Application example 2)

[1978] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1979] Conventional image generation systems require large amounts of computational resources to generate highly accurate images, and they have difficulty reflecting the specific emotional satisfaction desired by users in real time. Furthermore, there is a lack of a way to effectively utilize user feedback in the image generation process. As a result, images that do not fully meet the user's requirements are often generated, and satisfactory results cannot be achieved.

[1980] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for building a distributed network, means for supplying computing resources to each terminal, means for distributing image generation tasks, means for processing the distributed image generation tasks and transmitting generated images to the server, means for analyzing and selecting the generated images using AI, means for generating high-precision images using GAN based on the selected images, means for calculating rewards and paying the rewards in electronic money to terminals that contributed to the generation, means for collecting real-time emotional data from users, analyzing it, and reflecting it in the image generation process, and means for linking images generated via a terminal such as a smartphone with user emotional feedback. This makes it possible to generate high-precision images while also reflecting user emotional feedback in real time, thereby achieving image generation that greatly improves user satisfaction.

[1981] A "distributed network" is a network in which multiple devices are interconnected and can share and process data without centralized control.

[1982] "Computational resources" refers to the capabilities of hardware and software to perform computational processing, and specifically includes the performance of CPUs and GPUs.

[1983] "Image generation task" means a set of instructions for processing or computing required to generate a particular image.

[1984] A "server" is a central computer that processes and provides data in response to requests from multiple terminals within a network.

[1985] A "terminal" refers to a computer or device that is connected to a distributed network and provides computing resources.

[1986] "AI analysis and selection methods" refers to technology that uses artificial intelligence to evaluate collected images and select the most suitable ones.

[1987] "GAN" stands for generative adversarial network, a type of algorithm that generates realistic images by training two neural networks, a generator and a discriminator, to compete with each other.

[1988] "High-precision images" refer to detailed, high-quality images that closely resemble real-world scenes and objects.

[1989] "Electronic money" means currency provided in digital form, an asset that is exchanged electronically.

[1990] "Real-time emotional data" refers to information that collects the user's emotional state in real time and processes that data immediately.

[1991] "Emotional feedback" refers to the evaluation and response of a system based on emotional expressions and data provided by the user.

[1992] This invention is a system that uses a distributed network to generate highly accurate images, and also has the function of recognizing and reflecting user emotions in real time. The following is a specific embodiment of this invention.

[1993] Server Features

[1994] The server has the following main functions:

[1995] 1. Networking:

[1996] The server creates a distributed network and configures it so that each terminal can provide computing resources. It uses public IP addresses to provide an environment where terminals can easily connect.

[1997] 2. Task distribution:

[1998] The system divides image generation requests received from users into small tasks and distributes them to each device. For example, it gives detailed instructions such as "generate the texture of sand" or "generate the silhouette of a distant mountain" to each device.

[1999] 3. Image Collection and Analysis:

[2000] Images generated by each device are collected and analyzed and evaluated using AI algorithms, and the best images are selected based on evaluation criteria such as texture, color, and composition.

[2001] 4. Emotional engine regulation:

[2002] The system analyzes the user's emotional data in real time and reflects it at each stage of the generation process. This is achieved by analyzing the emotional information provided by the user through a camera or microphone.

[2003] 5. GAN Processing:

[2004] To generate the final high-precision image, a generative adversarial network (GAN) is used based on the selected images. GAN includes two neural networks: a generator and a discriminator.

[2005] 6. Reward Allocation:

[2006] The system checks the distributed network logs, calculates rewards based on the computing resources provided, and pays each device electronic money.

[2007] Device Features

[2008] The terminal has the following features:

[2009] 1. Registering with the Network:

[2010] They connect to a server, register with the decentralized network, provide computing resources, and receive tokens in return.

[2011] 2. Receiving and processing tasks:

[2012] It receives and processes image generation tasks from the server, and includes metadata such as the usage of computing resources when sending the generated images to the server.

[2013] User Roles

[2014] Users have the following roles:

[2015] 1. Image generation request:

[2016] Request the server to generate a specific image, specifying specific content such as a "mountain landscape" or a "city night view."

[2017] 2. Providing Emotion Data:

[2018] Cameras and sensors are used to provide facial and voice data in real time, which is analyzed by the emotion engine and reflected in image generation.

[2019] Specific examples

[2020] A user opens the app and requests that an image of a "summer beach sunset" be generated. The app collects the user's emotional data in real time via the camera and microphone. When the user expresses emotions such as "happy" or "moved" while looking at the image, that data is sent to the server and reflected in the image generation. Finally, an image that satisfies the user is generated and displayed on the smartphone.

[2021] An example prompt for a generative AI model is:

[2022] "Generate a summer seaside scene. I'd like the image to have a tranquil, relaxing atmosphere with the sound of the waves. Additionally, I'd like the sunset to be reflected on the water to enhance its beauty."

[2023] The above is an embodiment of the present invention. This system not only generates highly accurate images but also reflects real-time emotional feedback from the user, thereby improving user satisfaction.

[2024] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2025] Step 1:

[2026] Request Receipt

[2027] The server receives an image generation request from a user.

[2028] (Input) The user's request (e.g., "Summer sunset at the beach").

[2029] (Output) Initial settings for the image generation task.

[2030] (Specific operation) A user requests the generation of a specific image through the app, and the server receives the request, which may include a prompt for the generating AI model.

[2031] Step 2:

[2032] Task division

[2033] The server divides the image generation request into multiple subtasks.

[2034] (Input) The received image generation request.

[2035] (Output) A set of divided image generation tasks.

[2036] (Specific operation) The server breaks down the data into specific parts such as "generating the texture of sand" or "adjusting the color of the sunset" and converts each part into a processable format.

[2037] Step 3:

[2038] Task Distribution

[2039] The server distributes the divided tasks to the appropriate terminals.

[2040] (Input) A set of divided image generation tasks.

[2041] (Output) Tasks sent to each device.

[2042] (Specific operation) The server assigns tasks and sends instructions based on each terminal's available computing resources and past performance.

[2043] Step 4:

[2044] Receiving and processing tasks

[2045] The terminal processes the received task and sends the results to the server.

[2046] (Input) Image generation task sent from the server.

[2047] (Output) The generated subimage and metadata.

[2048] (Specific operation) Each device executes the received task, specifically generating the texture and color of a specific image, and sends this to the server along with metadata including the usage status of computing resources.

[2049] Step 5:

[2050] Image collection and analysis

[2051] The server collects images sent from each device and analyzes them using AI algorithms.

[2052] (Input) Partial images and metadata sent from each device.

[2053] (Output) The analyzed image data.

[2054] (Specific operation) The server evaluates the texture, color, composition, etc. of each image and selects the most suitable image using an AI algorithm.

[2055] Step 6:

[2056] Emotion data collection and analysis

[2057] The server collects and analyzes the user's real-time emotional data.

[2058] (Input) Emotion data from the user via camera and microphone.

[2059] (Output) Parsed emotion information.

[2060] (Specific operation) While the user is checking the generated image, facial expression and voice data is collected via the smartphone's camera and microphone, and the server analyzes this data using the emotion engine.

[2061] Step 7:

[2062] Reflecting and adjusting emotional data

[2063] The server adjusts image generation parameters based on the analyzed emotional information.

[2064] (Input) Analyzed emotion information.

[2065] (Output) Adjusted image generation parameters.

[2066] (Specific operation) The server adjusts the parameters of the generation process according to the user's positive or negative emotions. This adjustment ensures that the next task executed reflects the emotional data.

[2067] Step 8:

[2068] Final GAN ​​processing

[2069] The server generates the final image using GAN based on the selected images.

[2070] (Input) Selected partial images and adjusted parameters.

[2071] (Output) High-precision final image.

[2072] (Specific operation) Using a GAN (generative adversarial network) with a generator and discriminator, the final result is a highly accurate and realistic image.

[2073] Step 9:

[2074] Final image courtesy

[2075] The server provides the generated final image to the user.

[2076] (Input) High-precision final image.

[2077] (Output) The image that is displayed on the user device.

[2078] (Specific operation) The server sends the generated final image to the user's device, and the app displays it.

[2079] Step 10:

[2080] Reward Allocation

[2081] The server distributes rewards to the terminals that participate in the task processing.

[2082] (Input) Distributed network log and contribution data of each device.

[2083] (Output) Reward payment to each device.

[2084] (Specific operation) The server analyzes the logs of the distributed network, calculates rewards based on the degree of contribution, and pays electronic money to each terminal.

[2085] This concludes the explanation of the specific processing steps. This system enables highly accurate image generation and reflects the user's emotional feedback in real time.

[2086] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2087] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2088] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2089] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2090] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2091] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2092] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2093] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2094] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2095] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2096] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2097] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2098] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2099] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2100] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2101] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2102] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2103] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2104] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2105] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2106] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2107] The following is further disclosed regarding the above embodiment.

[2108] (Claim 1)

[2109] A means of building a decentralized network;

[2110] A means for supplying computing resources to each terminal;

[2111] means for distributing image generation tasks;

[2112] means for processing the distributed image generation tasks and transmitting the generated images to a server;

[2113] A means of analyzing and selecting the generated images using AI,

[2114] A method for generating highly accurate images using GAN based on the selected images;

[2115] A means for calculating a reward and paying the reward in electronic money to the terminal that contributed to the generation;

[2116] A system including:

[2117] (Claim 2)

[2118] The system of claim 1, wherein the means for analyzing and selecting the generated images using AI selects images based on evaluation criteria such as texture and color.

[2119] (Claim 3)

[2120] 10. The system of claim 1, further comprising means for checking a log of the distributed network and calculating a reward based on the provision of computational resources.

[2121] "Example 1"

[2122] (Claim 1)

[2123] A means of building a decentralized network;

[2124] A means for supplying computing resources to each terminal;

[2125] means for receiving and decomposing image generation requests;

[2126] A means for assigning image generation tasks to each terminal;

[2127] means for processing the distributed image generation tasks and transmitting the generated images to a server;

[2128] means for evaluating and selecting the generated images;

[2129] A means for generating a highly accurate image using a generative network based on the selected image;

[2130] A means for calculating a reward and paying the reward in electronic money to the terminal that contributed to the generation;

[2131] A system including:

[2132] (Claim 2)

[2133] The system of claim 1, wherein the means for analyzing and selecting the generated images using AI selects images based on evaluation criteria such as texture and color.

[2134] (Claim 3)

[2135] 10. The system of claim 1, further comprising means for checking a log of the distributed network and calculating a reward based on the provision of computational resources.

[2136] "Application Example 1"

[2137] (Claim 1)

[2138] A means of building a decentralized network;

[2139] A means for supplying computing resources to each terminal;

[2140] means for distributing image generation tasks;

[2141] means for processing the distributed image generation tasks and transmitting the generated images to a server;

[2142] A means of analyzing and selecting the generated images using AI,

[2143] A method for generating highly accurate images using GAN based on the selected images;

[2144] A means for calculating a reward and paying the reward in electronic money to the terminal that contributed to the generation;

[2145] means for using the generated image for product inspection and returning the inspection results to the robot;

[2146] A system including:

[2147] (Claim 2)

[2148] The system of claim 1, wherein the means for analyzing and selecting the generated images using AI selects images based on evaluation criteria such as texture and color, and inspects products for defects and abnormalities based on the images.

[2149] (Claim 3)

[2150] The system of claim 1, further comprising a means for checking the log of the distributed network and calculating rewards based on the provision of computing resources, and a means for performing quality control based on the inspection results of the product.

[2151] "Example 2: Combining Emotion Engines"

[2152] (Claim 1)

[2153] A means of building a decentralized network;

[2154] A means for supplying computing resources to each terminal;

[2155] means for distributing image generation tasks;

[2156] means for processing the distributed image generation tasks and transmitting the generated images to a server;

[2157] A means of analyzing and selecting the generated images using AI,

[2158] means for analyzing user emotion data and adjusting parameters for image generation;

[2159] A method for generating highly accurate images using GAN based on the selected images;

[2160] A means for calculating a reward and paying the reward in electronic money to the terminal that contributed to the generation;

[2161] A system including:

[2162] (Claim 2)

[2163] The system of claim 1, wherein the means for analyzing and selecting the generated images using AI selects images based on evaluation criteria such as texture and color.

[2164] (Claim 3)

[2165] 10. The system o...

Claims

1. A means of building a decentralized network; A means for supplying computing resources to each terminal; means for distributing image generation tasks; means for processing the distributed image generation tasks and transmitting the generated images to a server; A means of analyzing and selecting the generated images using AI, A method for generating highly accurate images using GAN based on the selected images; A means for calculating a reward and paying the reward in electronic money to the terminal that contributed to the generation; A system including:

2. The system according to claim 1, wherein the means for analyzing and selecting the generated images using AI selects images based on evaluation criteria such as texture and color.

3. The system of claim 1 , further comprising means for checking a log of the distributed network and calculating a reward based on the provision of computing resources.

Images