System

Abstract

A system is provided.SOLUTION: A system includes means for acquiring a handwriting sample by user input, means for extracting handwriting features by analyzing the acquired handwriting sample, means for learning and storing a handwriting model of the user based on the extracted handwriting features, means for receiving a character string by user input and generating an image by making the input character string resemble the user's handwriting using the stored handwriting model, and means for transmitting and displaying the generated image to a user terminal.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] In recent years, with the spread of smart devices, text input in digital environments has become common. However, digitized text does not reflect an individual's handwriting, and therefore tends to lose individuality and handwriting characteristics. In particular, there is a growing need to reproduce the individuality of handwritten characters digitally, but existing technologies do not allow for this to be achieved easily and with high accuracy. Therefore, there is a need to develop technology that can faithfully reproduce a user's handwriting and add individuality to digital text. [Means for solving the problem]

[0005] The present invention aims to provide a system that acquires a user's handwritten character sample, analyzes its handwriting characteristics, and generates a machine learning model. Then, by using this machine learning model to output a character string entered by the user as an image that resembles the user's handwriting, individual handwriting can be reproduced even in a digital environment. To this end, the present invention provides a system having the following means:

[0006] 1. A means for acquiring handwritten character samples input by the user as image data.

[0007] 2. A method for analyzing images of acquired handwritten character samples and extracting handwriting features.

[0008] 3. A means of learning and storing a model of the user's handwriting based on the extracted handwriting features using machine learning algorithms.

[0009] 4. A means for receiving a string of characters entered by a user and using the stored handwriting model to generate an image of the string of characters that resembles the user's handwriting.

[0010] 5. A means for transmitting the generated image to the user terminal and displaying it.

[0011] These methods allow users to easily generate character string images that faithfully reproduce their own handwriting, enabling them to express their individuality in digital communications.

[0012] "User input" refers to the act of a user providing text or information to a system.

[0013] "Handwritten character sample" refers to image data of characters handwritten by the user.

[0014] "Image data" refers to data that represents visual information in a digital format.

[0015] "Image analysis" is a technique that uses algorithms to analyze digital images and extract features and information.

[0016] "Handwriting features" refer to the characteristics of a person's handwritten characters, such as pressure, line thickness, angle, and stroke order.

[0017] A "machine learning model" refers to an algorithm or system that learns from data and makes predictions or classifications.

[0018] A "database" refers to a system that organizes and stores data so that it can be easily searched and managed.

[0019] A "character string" refers to text information consisting of multiple characters arranged in succession.

[0020] A "handwriting imitation image" refers to a digital image created to mimic the characteristics of handwritten characters.

[0021] A "user terminal" refers to an electronic device that can be directly operated by a user, such as a smartphone or tablet.

[0022] "Transmission" refers to the act of moving data or information from one point to another.

[0023] "Display" refers to the act of visually presenting digital information on a screen. [Brief explanation of the drawings]

[0024] [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

[0025] 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.

[0026] First, the terms used in the following description will be explained.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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."

[0032] [First embodiment]

[0033] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0034] 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.

[0035] 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).

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0041] 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.

[0042] 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.

[0043] 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.

[0044] 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."

[0045] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs a character string input by the user as an image that resembles the user's handwriting. A specific embodiment of the system will be described below.

[0046] 1. Acquiring handwriting data

[0047] Program processing:

[0048] A user writes characters by hand in an input field provided on a device (smartphone, tablet, etc.). The device captures the handwritten characters as image data and sends it to a server. This image data also includes the user's identification information.

[0049] Examples:

[0050] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0051] 2. Training the handwriting model

[0052] Program processing:

[0053] The server analyzes the received image data and extracts handwriting features, such as line thickness, pressure, curve angle, and stroke order. Based on the extracted handwriting features, a machine learning algorithm is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0054] Examples:

[0055] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0056] 3. Entering a string

[0057] Program processing:

[0058] The user inputs a desired character string into the terminal, which then transmits the input character string and the user's identification information to the server.

[0059] Examples:

[0060] The user enters the string "Hello" as text into the input field of the terminal, which then sends it to the server.

[0061] 4. Generating handwriting imitation images

[0062] Program processing:

[0063] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are often used to generate the image.

[0064] Examples:

[0065] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0066] 5. Displaying the generated results

[0067] Program processing:

[0068] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0069] Examples:

[0070] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0071] This system can faithfully reproduce the handwriting of individual users in a digital environment, adding individuality to digital text. This system can be used for a variety of purposes, including handwritten notes, letters, and electronic signatures.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The user writes characters by hand in an input field on the terminal.

[0075] The device captures the handwritten characters as image data.

[0076] Step 2:

[0077] The terminal transmits the captured handwritten character image and user identification information to the server.

[0078] Step 3:

[0079] The server analyzes the received image data and extracts handwriting features.

[0080] Specifically, it analyzes characteristics such as line thickness, pressure, angle, and stroke order.

[0081] Step 4:

[0082] The server uses a machine learning algorithm to generate a handwriting model specific to the user based on the extracted handwriting features.

[0083] Step 5:

[0084] The server stores the generated handwriting model in a database.

[0085] Step 6:

[0086] The user inputs a desired character string into an input field on the terminal.

[0087] Step 7:

[0088] The terminal transmits the input string and the user identification information to the server.

[0089] Step 8:

[0090] The server retrieves the corresponding handwriting model from the database based on the user identification information.

[0091] Step 9:

[0092] The server uses the acquired handwriting model to generate a handwriting imitation image that resembles the user's own handwriting for the input character string.

[0093] Step 10:

[0094] The server transmits the generated handwriting imitation image to the terminal.

[0095] Step 11:

[0096] The terminal displays the received handwriting imitation image on the user's screen.

[0097] Example 1

[0098] 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."

[0099] Conventional digital text generation methods have difficulty faithfully reproducing the handwriting of individual users, making it impossible to create unique digital content. Therefore, there is a demand for technology that can accurately reproduce a user's handwriting even in a digital environment and generate unique handwritten-style notes, letters, electronic signatures, etc.

[0100] 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.

[0101] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string that resembles the user's handwriting using the saved handwriting model, and means for transmitting the generated image to a user terminal and displaying it, thereby enabling the generation of digital text that accurately reproduces the user's handwriting.

[0102] "User input" refers to manual operations or data input by a user into the system.

[0103] A "handwritten character sample" is a digital image of a character handwritten by a user.

[0104] "Analysis" is the process of analyzing data and information in detail and extracting its meaning and characteristics.

[0105] "Handwriting features" are characteristic information about a particular handwriting, such as the thickness of the lines of handwritten characters, the pressure of the writing, the angle of the curves, and the order of strokes.

[0106] A "handwriting model" is a machine learning model that learns the user's unique handwriting characteristics.

[0107] "Learning" is the process by which machine learning algorithms understand patterns and features based on data.

[0108] "Storage" is the act of permanently recording data or information on a storage medium.

[0109] "Generating an image" means creating new image data using a specific algorithm.

[0110] A "user terminal" is a device (e.g., a smartphone, tablet, etc.) that a user uses to access the system.

[0111] "Display" refers to the act of visually presenting generated data or information to a user.

[0112] "Identification information" is data for uniquely identifying a user (for example, a user ID or authentication information).

[0113] "Image input means" refers to a device or function that captures a user's handwritten characters as a digital image.

[0114] "Image analysis" is the process of extracting information from digital images and analyzing it.

[0115] A "machine learning algorithm" is an algorithm that improves itself based on data.

[0116] A "database" is a system for efficiently managing, storing, and retrieving large amounts of data.

[0117] A "handwriting imitation image" is a digital image generated to resemble a user's own handwriting.

[0118] "Generation means" refers to the functions or devices that allow the system to create or convert new data or information.

[0119] A "prompt sentence" is text data that is input into an AI model.

[0120] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs an image of a character string input by the user that resembles the user's handwriting. Specific embodiments of the system are described below.

[0121] Acquiring handwriting data

[0122] A user writes characters by hand in an input field provided on a device such as a smartphone or tablet. When the user writes characters using a stylus pen or their finger, the device captures the characters as image data. The device then sends this image data and the user's identification information to a server. The hardware used is a typical smartphone or tablet, and the software has an image capture function built in.

[0123] Examples:

[0124] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0125] Training a handwriting model

[0126] The server analyzes the received image data and extracts handwriting features. This analysis uses an image processing library (e.g., OpenCV). Specifically, features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. Based on the extracted handwriting features, a machine learning algorithm (e.g., TENSORFLOW (registered trademark) or PyTorch) is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0127] Examples:

[0128] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0129] Entering a string

[0130] The user enters a desired string on the device. For example, they enter "hello" into a text input field. The device then sends the string and user identification information to the server. The hardware used includes a typical smartphone or tablet input field.

[0131] Examples:

[0132] The user enters the string "hello" as text into the terminal, which then sends it to the server.

[0133] Generating handwriting imitation images

[0134] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image.

[0135] Examples:

[0136] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0137] Viewing the generated results

[0138] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen, allowing the user to confirm that the input string appears similar to their own handwriting.

[0139] Examples:

[0140] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0141] Example prompt sentence:

[0142] "I would like to develop a system that acquires a user's handwriting, analyzes it using a machine learning model, and generates a handwritten-like image. Please tell me the specific steps: the user inputs characters into the device → the server analyzes the handwriting and generates a model → the input string is converted into a handwritten-like image and displayed. Please also tell me the technology and algorithms used."

[0143] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0144] Step 1:

[0145] The user writes handwritten characters on the device.

[0146] The user writes handwritten characters in an input field provided on the screen of a smartphone or tablet. The user inputs using a stylus pen or a finger. The device captures the handwritten characters as image data. The input data is an image of the handwritten characters, and the output data is the captured image file.

[0147] Specific behavior:

[0148] The user writes the words "Good morning" in the input field.

[0149] The device captures this text as an image and saves it in an internal temporary file.

[0150] Step 2:

[0151] The device captures the handwritten image and sends it to the server.

[0152] The device sends the captured image data and the user's identification information to the server. The input data is the image and user identification information acquired in the previous step, and the output data is the HTTP request sent to the server.

[0153] Specific behavior:

[0154] The device sends the captured image data of "Good morning" to the server as an HTTP POST request.

[0155] Step 3:

[0156] The server analyzes the image data and extracts handwriting features.

[0157] The server analyzes the received image data and uses an image processing library (e.g., OpenCV) to extract handwriting features. Features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. The input data is an image file, and the output data is the extracted handwriting feature data.

[0158] Specific behavior:

[0159] The server reads the "Good morning" image file and converts it to grayscale.

[0160] The server quantifies the thickness of the lines and the angle of the curves and extracts this as feature data.

[0161] Step 4:

[0162] The server uses machine learning algorithms to generate and store handwriting models.

[0163] The server uses the extracted handwriting feature data to generate a user-specific handwriting model using a machine learning algorithm (e.g., TensorFlow or PyTorch). The generated model is stored in a database. The input data is the handwriting feature data, and the output data is the trained handwriting model.

[0164] Specific behavior:

[0165] The server uses a neural network to learn the feature data and generate a handwriting model for the user.

[0166] Save the generated model in the database.

[0167] Step 5:

[0168] The user enters a string into a text input field on the device.

[0169] The user inputs a desired string in text format into the terminal. The input data is the text string and user identification information, and the output data is an HTTP request sent to the server.

[0170] Specific behavior:

[0171] The user enters the string "Hello" into the text input field on the device.

[0172] The terminal transmits this text data and the user identification information to the server.

[0173] Step 6:

[0174] The server generates a handwriting imitation image using the user's handwriting model.

[0175] The server retrieves the corresponding handwriting model from the database based on the user's identification information. Using the retrieved model, it generates an image that resembles the user's handwriting from the input string. GAN or other image generation algorithms are used for image generation. The input data are the text string and the handwriting model, and the output data is a handwriting imitation image.

[0176] Specific behavior:

[0177] The server reads the corresponding user's handwriting model from the database.

[0178] Using the text string "hello" as input, a handwriting imitation image is generated using the GAN algorithm.

[0179] Step 7:

[0180] The server generates a handwriting imitation image and sends it to the terminal, which displays it.

[0181] The server sends the generated handwriting imitation image to the terminal, which displays the received image on the user's screen. The input data is the handwriting imitation image, and the output data is the displayed image.

[0182] Specific behavior:

[0183] The server sends the generated handwriting imitation image of "hello" to the terminal as an HTTP response.

[0184] The terminal receives this image, places it on the display area, and presents it to the user.

[0185] (Application example 1)

[0186] 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."

[0187] In conventional virtual stores, it was difficult for users to input messages in their own handwriting and have them reflected in customized products. This made it difficult to provide custom-made products that reflected each user's individuality in a digital environment, posing challenges in improving user satisfaction. Furthermore, existing systems lack a method for faithfully reproducing a user's handwriting, and there is a lack of technology for creating handwritten-style customized products.

[0188] 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.

[0189] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string by using the saved handwriting model to resemble the user's handwriting, means for adding a message that looks like the user's handwriting to customized products in the virtual store, and means for transmitting the generated image to a user terminal and displaying it. This enables users to easily create and order customized products with original designs using their own handwriting.

[0190] "User input" includes characters and information entered by a user using a device.

[0191] A "handwriting sample" is a character written by a user in their own handwriting on a device.

[0192] "Handwriting characteristics" refer to characteristics such as the thickness of lines, writing pressure, and angle of curves that are specific to the user's handwritten characters.

[0193] A "handwriting model" is a digital model generated using a machine learning algorithm based on the user's handwriting characteristics.

[0194] "Image generation" is a process of generating an image that resembles the user's handwriting using a handwriting model based on input text.

[0195] A "virtual store" is a virtual store operated on the Internet where users can select and purchase products.

[0196] A "customized product" is a product that is individually created by a user to add specific elements (in this case, a handwritten message) to suit their preferences.

[0197] "User identification information" is information for identifying a specific user, and typically includes a user ID and authentication information.

[0198] The "image input means" is a means for capturing characters written by a user on a device as an image.

[0199] The "generation means" is a means for converting a character string input by a user into a handwriting imitation image using a specific algorithm.

[0200] The "display means" is a means for displaying the generated handwriting imitation image on the user terminal.

[0201] MODE FOR CARRYING OUT THE INVENTION

[0202] System Configuration

[0203] The system for implementing this invention consists of a series of processes: a user inputs handwritten characters on a terminal, a handwriting model is generated for the user based on the input, and finally, the user's handwritten-style message is added to a customized product in a virtual store. The specific system configuration is described below.

[0204] Hardware Configuration

[0205] User terminals include smartphones, tablets, PCs, etc. These terminals are equipped with image input means (cameras, touch screens, etc.).

[0206] The server used is a cloud server, which has the capacity and database to run machine learning algorithms.

[0207] Software Configuration

[0208] The overall system utilizes the following software:

[0209] 1. Image analysis software: Using image processing libraries such as OpenCV, we analyze the user's handwritten character samples and extract handwriting features.

[0210] 2. Machine learning algorithms: Using TensorFlow and PyTorch, we generate a handwriting model of the user based on the extracted handwriting features.

[0211] 3. Generative Adversarial Network (GAN): Using a GAN, an image of the input string is generated that resembles the user's handwriting based on a model of the user's own handwriting.

[0212] 4. Image editing software: Use PIL (Python Imaging Library) to overlay the generated handwriting imitation image onto the customized product image.

[0213] Program processing

[0214] The server captures handwriting samples entered by the user on the device and transmits them to the server. The captured image is analyzed on the server side to extract handwriting features. Based on the extracted features, a machine learning algorithm is used to generate a handwriting model of the user and store it in a database.

[0215] When a user wishes to create a customized product, a request is sent from the terminal along with a character string. The server searches for a handwriting model based on the user's identification information and converts the input character string into a handwriting imitation image. The generated handwriting imitation image is overlaid on the customized product using the PIL library, and finally sent to the user's terminal for display.

[0216] Specific examples

[0217] For example, when a user writes "thank you" by hand using a smartphone, the character is captured as an image and sent to the server. The server analyzes the image and extracts handwriting features. Based on this, a handwriting model of the user is generated and stored in a database.

[0218] When a user customizes a T-shirt in the virtual store, they input the word "thank you" and send the text data to the server. The server uses a model of the user's handwriting to convert the input text into a handwriting imitation image and overlay this image on the T-shirt design. Finally, the generated T-shirt design is displayed on the user's smartphone.

[0219] Prompt Sentence Examples

[0220] When a user makes a request to the server for a customized product, the prompt text can be written as follows:

[0221] User ID: 12345, Text: "Thank you"

[0222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0223] Step 1:

[0224] The user inputs handwritten characters on the device. The user writes characters such as "Thank you" by hand on the touch screen of a smartphone or tablet. This input is saved as image data on the device.

[0225] Input: Handwritten characters

[0226] Output: Image data

[0227] Step 2:

[0228] The device sends the captured image data to the server, along with the user's identification information, and uploads the image data to the server via the Internet.

[0229] Input: Image data, user identification information

[0230] Output: None (data transfer to server)

[0231] Step 3:

[0232] The server analyzes the received image data and extracts handwriting features, such as line thickness, pen pressure, and curve angle, using an image processing library such as OpenCV, to extract numerical data.

[0233] Input: Image data

[0234] Output: Handwriting features (numerical data)

[0235] Step 4:

[0236] The server uses a machine learning algorithm to generate a handwriting model of the user based on the extracted handwriting features and stores it in a database. TensorFlow and PyTorch are used to learn the handwriting features and generate a model specific to the user.

[0237] Input: Handwriting features

[0238] Output: Handwritten model (model data)

[0239] Step 5:

[0240] When a user creates a customized product in the virtual store, the user inputs an input string (e.g., "custom message") into the terminal and transmits this data to the server. The terminal then transmits the text data entered by the user and the user identification information to the server.

[0241] Input: Text data (input string), user identification information

[0242] Output: None (data transfer to server)

[0243] Step 6:

[0244] The server searches for and retrieves the corresponding handwriting model from the database based on the user identification information, searches for the saved handwriting model, and loads the corresponding model into memory.

[0245] Input: User Identification Information

[0246] Output: Handwritten model (model data)

[0247] Step 7:

[0248] The server uses a handwriting model to generate an image that resembles the user's handwriting. A generative adversarial network (GAN) is used to convert the input text into a handwriting-imitating image.

[0249] Input: Text data, handwriting model

[0250] Output: Handwriting imitation image

[0251] Step 8:

[0252] The server overlays the generated handwriting imitation image onto the customized product image. Using PIL, the handwriting imitation image is synthesized onto the customized product design image.

[0253] Input: Handwriting imitation image, customized product image

[0254] Output: Customized product image (after compositing)

[0255] Step 9:

[0256] The server transmits the synthesized customized product image to the user terminal, and the generated customized product image is transmitted to the user terminal via the Internet and displayed.

[0257] Input: Customized product image (after compositing)

[0258] Output: None (data transfer to user terminal)

[0259] Step 10:

[0260] The user terminal displays the received customized product image to the user, and the image is displayed on the terminal display so that the user can check the product customized with their own handwriting.

[0261] Input: Customized product image (after receiving)

[0262] Output: User confirms customized product image

[0263] 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.

[0264] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[0265] 1. Acquisition of handwriting data and emotion data

[0266] Program processing:

[0267] The user writes characters by hand in an input field on the device. The device captures the handwritten characters as image data. The device also simultaneously captures the user's voice and facial expressions while writing using a camera and microphone. The acquired voice and facial expression data are sent to an emotion engine, which analyzes the user's emotional state. The image data and emotional state information are then sent to the server.

[0268] Examples:

[0269] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[0270] 2. Learning handwriting and emotion models

[0271] Program processing:

[0272] The server analyzes the received image data and extracts handwriting features, such as pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate a handwriting model and emotion model specific to the user and stores them in a database.

[0273] Examples:

[0274] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[0275] 3. Entering text and emotions

[0276] Program processing:

[0277] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[0278] Examples:

[0279] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[0280] 4. Generating handwriting imitation images

[0281] Program processing:

[0282] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[0283] Examples:

[0284] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[0285] 5. Displaying the generated results

[0286] Program processing:

[0287] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0288] Examples:

[0289] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[0290] This invention allows us to faithfully reproduce the handwriting and emotional state of individual users in a digital environment, thereby adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, etc.

[0291] The processing flow will be explained below.

[0292] Step 1:

[0293] The user writes characters by hand in an input field on the terminal.

[0294] The device captures the handwritten characters as image data.

[0295] Step 2:

[0296] The terminal transmits the captured handwritten character image and user identification information to the server.

[0297] Step 3:

[0298] The device captures the user's voice and facial expressions through a camera and microphone and sends them to the emotion engine, which analyzes the user's emotional state and sends the results to the server.

[0299] Step 4:

[0300] The server analyzes the received image data and extracts handwriting characteristics, such as line thickness, pressure, angle, and stroke order.

[0301] Step 5:

[0302] The server generates an emotion model based on the emotion state information received from the emotion engine, and the emotion model includes adjustment parameters for handwriting features according to the user's emotion state.

[0303] Step 6:

[0304] Based on the extracted handwriting features and emotion model, the server uses a machine learning algorithm to generate a handwriting model specific to the user and stores it in a database.

[0305] Step 7:

[0306] The user inputs a desired character string on the terminal.

[0307] The terminal transmits the character string, the user identification information, and the emotional state to the server.

[0308] Step 8:

[0309] The server retrieves a corresponding handwriting model from the database based on the user identification information, and further retrieves an emotion model based on the emotion state information.

[0310] Step 9:

[0311] The server uses the acquired handwriting model and emotion model to generate a handwriting imitation image of the input string that resembles the user's own handwriting. The style and angle of the characters are adjusted depending on the user's emotional state.

[0312] Step 10:

[0313] The server transmits the generated handwriting imitation image to the terminal.

[0314] Step 11:

[0315] The terminal displays the received handwriting imitation image on the user's screen.

[0316] For example, when a user writes "good morning" on a device, the device captures the characters as image data and sends it to the server. If the server recognizes that the user is in a "joy" emotional state, it performs analysis based on this information. When the user then types "hello," the server generates a handwriting imitation image with slightly thicker lines and softer curves, in line with the "joy" emotional state, and displays it on the device.

[0317] This invention makes it possible to generate character images that reflect the handwriting and emotions of individual users even in a digital environment, thereby imparting individuality and emotion to digital text. This opens up a wide range of applications, including handwritten-style notes, letters, and electronic signatures.

[0318] Example 2

[0319] 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."

[0320] Conventional systems can reproduce the characteristics of a user's handwriting when digitizing the user's handwritten text, but they are unable to reflect the user's emotional state. This makes it difficult to impart the user's personality and emotions to the digital text, resulting in a loss of the warmth and emotional expression inherent in handwritten text. The present invention aims to solve this problem by faithfully reproducing the user's handwriting and emotions in a digital environment.

[0321] 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.

[0322] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for analyzing the extracted handwriting features and emotional state to generate and save a handwriting model and emotional model dedicated to the user, means for receiving a character string input by the user and the emotional state and using the saved handwriting model and emotional model to generate an image that resembles the input character string and reflects the user's emotional state, and means for transmitting the generated image to a user terminal and displaying it. This makes it possible to generate a handwriting imitation image that reflects the user's individuality and emotions even in a digital environment.

[0323] The "means for acquiring handwritten character samples input by a user" refers to a device or method for capturing handwritten characters input by a user using a touch screen or a stylus pen and acquiring them as digital data.

[0324] The "means for extracting handwriting characteristics" refers to a device or method that analyzes and extracts characteristics such as writing pressure, line thickness, angle, and curve from an acquired handwritten character sample.

[0325] "Means for capturing emotional data" refers to a device or method that uses a camera or microphone to capture the user's facial expressions and voice and obtain them as digital data.

[0326] "Means for analyzing emotional state" refers to a device or method that analyzes captured facial expression and voice data to determine the emotional state of the user.

[0327] "Means for generating and storing handwriting models and emotion models" refers to a device or method that performs machine learning based on extracted handwriting features and analyzed emotional states, generates handwriting models and emotion models specific to the user, and stores them in a database.

[0328] The "means for generating an image that resembles the user's handwriting from an input character string and reflects the user's emotional state" refers to a device or method for generating a handwriting imitation image that resembles the user's handwriting, based on the character string and emotional state input by the user, using a stored handwriting model and emotional model.

[0329] The "means for transmitting the generated image to the user terminal and displaying it" refers to a device or method for transmitting the generated handwriting imitation image to the user terminal and displaying it on the screen of the terminal.

[0330] MODE FOR CARRYING OUT THE INVENTION

[0331] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[0332] Acquiring handwriting and emotion data

[0333] The user writes characters by hand in the input field on the device. The device captures this handwritten character as image data. Specifically, the device captures the user writing "good morning" on the touch screen with a stylus pen or finger and saves it as image data. The device also simultaneously captures the user's facial expressions and voice as they write using a camera and microphone. The acquired voice and facial expression data is sent to an emotion engine, which analyzes the user's emotional state. The device then sends this data to a server.

[0334] Examples:

[0335] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[0336] Learning handwriting and emotion models

[0337] The server analyzes the received image data and extracts handwriting features, specifically pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate handwriting and emotion models specific to the user and stores them in a database. Machine learning libraries used include TensorFlow and PyTorch.

[0338] Examples:

[0339] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[0340] String and emotion input

[0341] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[0342] Examples:

[0343] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[0344] Generating handwriting imitation images

[0345] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[0346] Examples:

[0347] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[0348] Viewing the generated results

[0349] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0350] Examples:

[0351] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[0352] Example prompt sentence:

[0353] A user writes "Good morning" in an input field by hand, and their voice (cheerful voice) and facial expression (happy smile) are captured to analyze their emotional state. Next, the same user types "Hello" and selects "Happy" as their emotional state. Using this information, the server should generate a handwriting imitation image of "Hello."

[0354] By using these steps, we can faithfully reproduce the user's handwriting and emotional state in a digital environment, adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, and more.

[0355] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0356] Step 1: User inputs a sample of their own handwriting

[0357] The user writes characters by hand using a stylus pen or their finger in the input field of the device. The device acquires this input and saves it as digital image data. A specific input is a string of characters such as "Good morning." The output is image data of the string of characters.

[0358] Step 2: Capture handwriting and emotion data

[0359] The device uses a camera and microphone to capture the user's facial expressions and voice while they are writing. The camera and microphone signals are processed to obtain facial expression data and voice data. Specifically, the device captures the user's smile and cheerful voice while they are writing "Good morning." The output is facial expression data and voice data.

[0360] Step 3: Analyze emotional state

[0361] The device sends the captured facial expression and voice data to the emotion engine, which analyzes the data and determines the user's emotional state. The specific inputs are facial expression and voice data, and the output is emotional state information such as "joy."

[0362] Step 4: Sending data

[0363] The terminal transmits image data of the handwritten characters and the analyzed emotional state to the server. The specific inputs are the image data and the emotional state information, and the output is the data arriving at the server.

[0364] Step 5: Extracting handwriting features

[0365] The server analyzes the received image data and extracts handwriting features such as pressure, line thickness, angle, and curve. Specifically, it uses an image processing algorithm to extract these features from the input data (handwriting image). The output is the extracted handwriting feature data.

[0366] Step 6: Generate emotion and handwriting models

[0367] The server applies machine learning algorithms to the extracted handwriting features and emotional state information to generate a user-specific handwriting model and emotion model. Specifically, it uses machine learning libraries such as TensorFlow and PyTorch. The input is the handwriting feature data and emotional state information, and the output is the generated model.

[0368] Step 7: Save the model

[0369] The server stores the generated handwriting and emotion models in a database. Specifically, it serializes the models and writes them to the database. The input is the model data, and the output is the results stored in the database.

[0370] Step 8: Entering Strings and Emotional States

[0371] The user inputs a desired string (e.g., "hello") and emotional state on the terminal. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The input is the string and emotional state information. The output is the input data on the terminal.

[0372] Step 9: Sending data (input text and emotional state)

[0373] The terminal sends the input string and emotional state to the server. The input is the string and emotional state information, and the output is the data arriving at the server.

[0374] Step 10: Getting the Model

[0375] The server retrieves the corresponding handwriting model and emotion model from the database. The input is the user identification information, and the output is the model data.

[0376] Step 11: Generate handwriting imitation images

[0377] The server uses the acquired model to generate a handwriting imitation image that matches the user's handwriting and emotional state. Specifically, it uses a generative adversarial network (GAN). The input is the string of characters, emotional state information, and model data, and the output is a handwriting imitation image.

[0378] Step 12: Sending the generated results

[0379] The server sends the generated handwriting imitation image to the terminal. The input is the handwriting imitation image, and the output is the image data arriving at the terminal.

[0380] Step 13: Display on the screen

[0381] The terminal displays the received handwriting imitation image on the user's screen. The input is the handwriting imitation image data, and the output is the display result on the user's screen.

[0382] (Application example 2)

[0383] 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."

[0384] Conventional systems generate handwriting imitation images based only on a user's handwriting samples, making it impossible to reflect changes in emotions or mood. Furthermore, there was a lack of a way to visually convey individual customers' emotions and personalities when improving customer experiences and promoting them in physical stores. In particular, there were limited ways for customers to share their emotions and feedback with other customers in the store. This made it difficult to provide an engaging customer experience using digital signage and displays.

[0385] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0386] In this invention, the server includes means for acquiring handwritten character samples and emotional states input by a user, means for analyzing the acquired handwritten character samples and emotional states to extract handwriting features and emotional features, and means for learning and saving a handwriting model and emotional model of the user based on the extracted handwriting features and emotional features, thereby enabling the generation of handwriting imitation images that reflect the user's individuality and emotions.

[0387] A "handwritten character sample" is a sample of characters input by handwriting by the user.

[0388] "Emotional state" is information that indicates the user's emotional or psychological state.

[0389] "Handwriting characteristics" refers to characteristics such as the shape, writing pressure, line thickness, angle, and curve of the user's handwritten characters.

[0390] "Emotional features" refer to features such as facial expressions, voice, and writing style based on the user's emotional state.

[0391] A "handwriting model" is a model that reproduces a user's unique handwriting, generated by a machine learning algorithm based on a user's handwriting samples.

[0392] The "emotion model" is a model for analyzing the user's emotional state and generating a handwriting imitation image that reflects the emotion.

[0393] "Image input means" refers to means for obtaining handwritten character samples using a camera, scanner, or the like.

[0394] The "emotion analysis means" is a means for analyzing voice and facial expression data to identify the emotional state of the user.

[0395] The "generating means" is a means for converting a character string input by a user into a handwriting imitation image using the stored handwriting model and emotion model.

[0396] The "database" is a storage device that stores the generated handwriting model and emotion model and makes them searchable as needed.

[0397] "User identification information" is information for uniquely identifying a user.

[0398] A "handwriting imitation image" is an image that reflects the user's handwriting and emotions, generated using a handwriting model and emotion model of the user.

[0399] Specific embodiments for carrying out the present invention will be described below.

[0400] The present invention is a system that generates handwritten-style message images based on a customer's handwriting sample and emotional state and displays them on an in-store display to improve customer experience in brick-and-mortar stores. This system generates a handwriting model and an emotion model based on user input data and uses them to generate a handwriting imitation image.

[0401] Hardware Configuration

[0402] Input device: Use an iPad (registered trademark) or an ANDROID (registered trademark) tablet, which allows customers to input handwritten messages.

[0403] Camera: Uses the device's built-in camera, which allows capturing customer facial expressions and analyzing their emotional state.

[0404] Microphone: Uses the device's built-in microphone, which allows us to capture the customer's voice and analyze their emotional state.

[0405] Server: Use a cloud server (e.g., AWS (registered trademark), Google (registered trademark) Cloud) to process large amounts of data and generate models.

[0406] Display: A large display will be used in the store to display the generated handwritten message image.

[0407] Software Configuration

[0408] Handwriting capture: The device app combines a journal and note-taking app with the ability to accept handwriting input from customers.

[0409] Emotion analysis engine: Uses software (e.g., FaceAPI, Azure® Cognitive Services) to capture voice and facial expression data and analyze emotions.

[0410] Data transmission: It has network functionality to transmit captured data to a cloud server in real time.

[0411] Model generation: On a cloud server, a handwriting model and an emotion model of the user are generated using a generative adversarial network (GAN) or a machine learning algorithm (e.g., TensorFlow, PyTorch).

[0412] Image generation & display: The generated handwriting imitation image is sent to the display via a web service and displayed in real time.

[0413] Specific examples

[0414] 1. Customer operations:

[0415] He writes in his own handwriting on a tablet inside the store, "I love this store!"

[0416] Cameras and microphones capture customers' facial expressions and voices.

[0417] 2. Emotion analysis:

[0418] The emotion analysis engine analyzes facial expressions and voice data to detect the emotional state of "joy."

[0419] 3. Data transmission:

[0420] The handwritten character sample and emotion data are sent to a cloud server.

[0421] 4. Model generation and image output:

[0422] The server analyzes handwriting features and emotion features based on the received data and generates a handwriting model and emotion model.

[0423] The sentence "I love this store!" is converted into a handwritten style that matches the emotion of joy, and is generated in slightly thicker strokes.

[0424] 5. Results display:

[0425] The generated handwriting imitation image is displayed on a large display inside the store.

[0426] The display features a hand-drawn image that reflects the customer's joyful emotions and reads, "We love this store!", for all customers to see.

[0427] Example prompts for generative AI models

[0428] "Generate a handwritten message image based on the handwritten characters and emotion data below.

[0429] Text: I love this store!

[0430] Emotion: Joy

[0431] Format: Handwritten, light background, thick strokes

[0432] This system will improve the customer experience in physical stores and make them feel more connected to other customers.

[0433] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0434] Step 1:

[0435] The device captures the user's handwritten text in the input field as image data, while the device's built-in camera and microphone simultaneously capture the user's facial expressions and voice, providing data for emotion analysis.

[0436] Input: Handwritten text, facial expression data, voice data

[0437] Output: Image data (handwritten characters), emotion analysis data (facial expressions, voice)

[0438] Step 2:

[0439] The device transmits the captured image data and sentiment analysis data to a server, which applies an image analysis algorithm to extract handwriting features from the image data and uses a sentiment analysis engine to extract sentiment features from the sentiment analysis data.

[0440] Input: Image data, emotion analysis data

[0441] Output: handwriting features, emotion features

[0442] Step 3:

[0443] The server uses the extracted handwriting features and emotion features to generate a user-specific handwriting model and emotion model using a machine learning algorithm (e.g., TensorFlow, PyTorch), and stores this in a database.

[0444] Input: handwriting features, emotion features

[0445] Output: Handwriting model, emotion model

[0446] Step 4:

[0447] The user inputs a desired character string and their emotional state on the terminal. The terminal transmits the data to the server. The server uses the stored handwriting model and emotion model to generate a handwriting imitation image that resembles the user's handwriting and corresponds to their emotional state.

[0448] Input: Input string, emotional state

[0449] Output: Handwriting imitation image

[0450] Step 5:

[0451] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen and also sends it to a large display in the store to share with other customers.

[0452] Input: Handwriting imitation image

[0453] Output: Images displayed on user devices and large displays

[0454] 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.

[0455] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0456] 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.

[0457] [Second embodiment]

[0458] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0459] 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.

[0460] 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).

[0461] 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.

[0462] 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.

[0463] 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).

[0464] 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. 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.

[0465] 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.

[0466] 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.

[0467] 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.

[0468] 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.

[0469] 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."

[0470] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs a character string input by the user as an image that resembles the user's handwriting. A specific embodiment of the system will be described below.

[0471] 1. Acquiring handwriting data

[0472] Program processing:

[0473] A user writes characters by hand in an input field provided on a device (smartphone, tablet, etc.). The device captures the handwritten characters as image data and sends it to a server. This image data also includes the user's identification information.

[0474] Examples:

[0475] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0476] 2. Training the handwriting model

[0477] Program processing:

[0478] The server analyzes the received image data and extracts handwriting features, such as line thickness, pressure, curve angle, and stroke order. Based on the extracted handwriting features, a machine learning algorithm is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0479] Examples:

[0480] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0481] 3. Entering a string

[0482] Program processing:

[0483] The user inputs a desired character string into the terminal, which then transmits the input character string and the user's identification information to the server.

[0484] Examples:

[0485] The user enters the string "Hello" as text into the input field of the terminal, which then sends it to the server.

[0486] 4. Generating handwriting imitation images

[0487] Program processing:

[0488] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are often used to generate the image.

[0489] Examples:

[0490] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0491] 5. Displaying the generated results

[0492] Program processing:

[0493] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0494] Examples:

[0495] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0496] This system can faithfully reproduce the handwriting of individual users in a digital environment, adding individuality to digital text. This system can be used for a variety of purposes, including handwritten notes, letters, and electronic signatures.

[0497] The processing flow will be explained below.

[0498] Step 1:

[0499] The user writes characters by hand in an input field on the terminal.

[0500] The device captures the handwritten characters as image data.

[0501] Step 2:

[0502] The terminal transmits the captured handwritten character image and user identification information to the server.

[0503] Step 3:

[0504] The server analyzes the received image data and extracts handwriting features.

[0505] Specifically, it analyzes characteristics such as line thickness, pressure, angle, and stroke order.

[0506] Step 4:

[0507] The server uses a machine learning algorithm to generate a handwriting model specific to the user based on the extracted handwriting features.

[0508] Step 5:

[0509] The server stores the generated handwriting model in a database.

[0510] Step 6:

[0511] The user inputs a desired character string into an input field on the terminal.

[0512] Step 7:

[0513] The terminal transmits the input string and the user identification information to the server.

[0514] Step 8:

[0515] The server retrieves the corresponding handwriting model from the database based on the user identification information.

[0516] Step 9:

[0517] The server uses the acquired handwriting model to generate a handwriting imitation image that resembles the user's own handwriting for the input character string.

[0518] Step 10:

[0519] The server transmits the generated handwriting imitation image to the terminal.

[0520] Step 11:

[0521] The terminal displays the received handwriting imitation image on the user's screen.

[0522] Example 1

[0523] 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."

[0524] Conventional digital text generation methods have difficulty faithfully reproducing the handwriting of individual users, making it impossible to create unique digital content. Therefore, there is a demand for technology that can accurately reproduce a user's handwriting even in a digital environment and generate unique handwritten-style notes, letters, electronic signatures, etc.

[0525] 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.

[0526] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string that resembles the user's handwriting using the saved handwriting model, and means for transmitting the generated image to a user terminal and displaying it, thereby enabling the generation of digital text that accurately reproduces the user's handwriting.

[0527] "User input" refers to manual operations or data input by a user into the system.

[0528] A "handwritten character sample" is a digital image of a character handwritten by a user.

[0529] "Analysis" is the process of analyzing data and information in detail and extracting its meaning and characteristics.

[0530] "Handwriting features" are characteristic information about a particular handwriting, such as the thickness of the lines of handwritten characters, the pressure of the writing, the angle of the curves, and the order of strokes.

[0531] A "handwriting model" is a machine learning model that learns the user's unique handwriting characteristics.

[0532] "Learning" is the process by which machine learning algorithms understand patterns and features based on data.

[0533] "Storage" is the act of permanently recording data or information on a storage medium.

[0534] "Generating an image" means creating new image data using a specific algorithm.

[0535] A "user terminal" is a device (e.g., a smartphone, tablet, etc.) that a user uses to access the system.

[0536] "Display" refers to the act of visually presenting generated data or information to a user.

[0537] "Identification information" is data for uniquely identifying a user (for example, a user ID or authentication information).

[0538] "Image input means" refers to a device or function that captures a user's handwritten characters as a digital image.

[0539] "Image analysis" is the process of extracting information from digital images and analyzing it.

[0540] A "machine learning algorithm" is an algorithm that improves itself based on data.

[0541] A "database" is a system for efficiently managing, storing, and retrieving large amounts of data.

[0542] A "handwriting imitation image" is a digital image generated to resemble a user's own handwriting.

[0543] "Generation means" refers to the functions or devices that allow the system to create or convert new data or information.

[0544] A "prompt sentence" is text data that is input into an AI model.

[0545] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs an image of a character string input by the user that resembles the user's handwriting. Specific embodiments of the system are described below.

[0546] Acquiring handwriting data

[0547] A user writes characters by hand in an input field provided on a device such as a smartphone or tablet. When the user writes characters using a stylus pen or their finger, the device captures the characters as image data. The device then sends this image data and the user's identification information to a server. The hardware used is a typical smartphone or tablet, and the software has an image capture function built in.

[0548] Examples:

[0549] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0550] Training a handwriting model

[0551] The server analyzes the received image data and extracts handwriting features. This analysis uses an image processing library (e.g., OpenCV). Specifically, features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. Based on the extracted handwriting features, a machine learning algorithm (e.g., TensorFlow or PyTorch) is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0552] Examples:

[0553] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0554] Entering a string

[0555] The user enters a desired string on the device. For example, they enter "hello" into a text input field. The device then sends the string and user identification information to the server. The hardware used includes a typical smartphone or tablet input field.

[0556] Examples:

[0557] The user enters the string "hello" as text into the terminal, which then sends it to the server.

[0558] Generating handwriting imitation images

[0559] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image.

[0560] Examples:

[0561] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0562] Viewing the generated results

[0563] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen, allowing the user to confirm that the input string appears similar to their own handwriting.

[0564] Examples:

[0565] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0566] Example prompt sentence:

[0567] "I would like to develop a system that acquires a user's handwriting, analyzes it using a machine learning model, and generates a handwritten-like image. Please tell me the specific steps: the user inputs characters into the device → the server analyzes the handwriting and generates a model → the input string is converted into a handwritten-like image and displayed. Please also tell me the technology and algorithms used."

[0568] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0569] Step 1:

[0570] The user writes handwritten characters on the device.

[0571] The user writes handwritten characters in an input field provided on the screen of a smartphone or tablet. The user inputs using a stylus pen or a finger. The device captures the handwritten characters as image data. The input data is an image of the handwritten characters, and the output data is the captured image file.

[0572] Specific behavior:

[0573] The user writes the words "Good morning" in the input field.

[0574] The device captures this text as an image and saves it in an internal temporary file.

[0575] Step 2:

[0576] The device captures the handwritten image and sends it to the server.

[0577] The device sends the captured image data and the user's identification information to the server. The input data is the image and user identification information acquired in the previous step, and the output data is the HTTP request sent to the server.

[0578] Specific behavior:

[0579] The device sends the captured image data of "Good morning" to the server as an HTTP POST request.

[0580] Step 3:

[0581] The server analyzes the image data and extracts handwriting features.

[0582] The server analyzes the received image data and uses an image processing library (e.g., OpenCV) to extract handwriting features. Features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. The input data is an image file, and the output data is the extracted handwriting feature data.

[0583] Specific behavior:

[0584] The server reads the "Good morning" image file and converts it to grayscale.

[0585] The server quantifies the thickness of the lines and the angle of the curves and extracts this as feature data.

[0586] Step 4:

[0587] The server uses machine learning algorithms to generate and store handwriting models.

[0588] The server uses the extracted handwriting feature data to generate a user-specific handwriting model using a machine learning algorithm (e.g., TensorFlow or PyTorch). The generated model is stored in a database. The input data is the handwriting feature data, and the output data is the trained handwriting model.

[0589] Specific behavior:

[0590] The server uses a neural network to learn the feature data and generate a handwriting model for the user.

[0591] Save the generated model in the database.

[0592] Step 5:

[0593] The user enters a string into a text input field on the device.

[0594] The user inputs a desired string in text format into the terminal. The input data is the text string and user identification information, and the output data is an HTTP request sent to the server.

[0595] Specific behavior:

[0596] The user enters the string "Hello" into the text input field on the device.

[0597] The terminal transmits this text data and the user identification information to the server.

[0598] Step 6:

[0599] The server generates a handwriting imitation image using the user's handwriting model.

[0600] The server retrieves the corresponding handwriting model from the database based on the user's identification information. Using the retrieved model, it generates an image that resembles the user's handwriting from the input string. GAN or other image generation algorithms are used for image generation. The input data are the text string and the handwriting model, and the output data is a handwriting imitation image.

[0601] Specific behavior:

[0602] The server reads the corresponding user's handwriting model from the database.

[0603] Using the text string "hello" as input, a handwriting imitation image is generated using the GAN algorithm.

[0604] Step 7:

[0605] The server generates a handwriting imitation image and sends it to the terminal, which displays it.

[0606] The server sends the generated handwriting imitation image to the terminal, which displays the received image on the user's screen. The input data is the handwriting imitation image, and the output data is the displayed image.

[0607] Specific behavior:

[0608] The server sends the generated handwriting imitation image of "hello" to the terminal as an HTTP response.

[0609] The terminal receives this image, places it on the display area, and presents it to the user.

[0610] (Application example 1)

[0611] 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."

[0612] In conventional virtual stores, it was difficult for users to input messages in their own handwriting and have them reflected in customized products. This made it difficult to provide custom-made products that reflected each user's individuality in a digital environment, posing challenges in improving user satisfaction. Furthermore, existing systems lack a method for faithfully reproducing a user's handwriting, and there is a lack of technology for creating handwritten-style customized products.

[0613] 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.

[0614] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string by using the saved handwriting model to resemble the user's handwriting, means for adding a message that looks like the user's handwriting to customized products in the virtual store, and means for transmitting the generated image to a user terminal and displaying it. This enables users to easily create and order customized products with original designs using their own handwriting.

[0615] "User input" includes characters and information entered by a user using a device.

[0616] A "handwriting sample" is a character written by a user in their own handwriting on a device.

[0617] "Handwriting characteristics" refer to characteristics such as the thickness of lines, writing pressure, and angle of curves that are specific to the user's handwritten characters.

[0618] A "handwriting model" is a digital model generated using a machine learning algorithm based on the user's handwriting characteristics.

[0619] "Image generation" is a process of generating an image that resembles the user's handwriting using a handwriting model based on input text.

[0620] A "virtual store" is a virtual store operated on the Internet where users can select and purchase products.

[0621] A "customized product" is a product that is individually created by a user to add specific elements (in this case, a handwritten message) to suit their preferences.

[0622] "User identification information" is information for identifying a specific user, and typically includes a user ID and authentication information.

[0623] The "image input means" is a means for capturing characters written by a user on a device as an image.

[0624] The "generation means" is a means for converting a character string input by a user into a handwriting imitation image using a specific algorithm.

[0625] The "display means" is a means for displaying the generated handwriting imitation image on the user terminal.

[0626] MODE FOR CARRYING OUT THE INVENTION

[0627] System Configuration

[0628] The system for implementing this invention consists of a series of processes: a user inputs handwritten characters on a terminal, a handwriting model is generated for the user based on the input, and finally, the user's handwritten-style message is added to a customized product in a virtual store. The specific system configuration is described below.

[0629] Hardware Configuration

[0630] User terminals include smartphones, tablets, PCs, etc. These terminals are equipped with image input means (cameras, touch screens, etc.).

[0631] The server used is a cloud server, which has the capacity and database to run machine learning algorithms.

[0632] Software Configuration

[0633] The overall system utilizes the following software:

[0634] 1. Image analysis software: Using image processing libraries such as OpenCV, we analyze the user's handwritten character samples and extract handwriting features.

[0635] 2. Machine learning algorithms: Using TensorFlow and PyTorch, we generate a handwriting model of the user based on the extracted handwriting features.

[0636] 3. Generative Adversarial Network (GAN): Using a GAN, an image of the input string is generated that resembles the user's handwriting based on a model of the user's own handwriting.

[0637] 4. Image editing software: Use PIL (Python Imaging Library) to overlay the generated handwriting imitation image onto the customized product image.

[0638] Program processing

[0639] The server captures handwriting samples entered by the user on the device and transmits them to the server. The captured image is analyzed on the server side to extract handwriting features. Based on the extracted features, a machine learning algorithm is used to generate a handwriting model of the user and store it in a database.

[0640] When a user wishes to create a customized product, a request is sent from the terminal along with a character string. The server searches for a handwriting model based on the user's identification information and converts the input character string into a handwriting imitation image. The generated handwriting imitation image is overlaid on the customized product using the PIL library, and finally sent to the user's terminal for display.

[0641] Specific examples

[0642] For example, when a user writes "thank you" by hand using a smartphone, the character is captured as an image and sent to the server. The server analyzes the image and extracts handwriting features. Based on this, a handwriting model of the user is generated and stored in a database.

[0643] When a user customizes a T-shirt in the virtual store, they input the word "thank you" and send the text data to the server. The server uses a model of the user's handwriting to convert the input text into a handwriting imitation image and overlay this image on the T-shirt design. Finally, the generated T-shirt design is displayed on the user's smartphone.

[0644] Prompt Sentence Examples

[0645] When a user makes a request to the server for a customized product, the prompt text can be written as follows:

[0646] User ID: 12345, Text: "Thank you"

[0647] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0648] Step 1:

[0649] The user inputs handwritten characters on the device. The user writes characters such as "Thank you" by hand on the touch screen of a smartphone or tablet. This input is saved as image data on the device.

[0650] Input: Handwritten characters

[0651] Output: Image data

[0652] Step 2:

[0653] The device sends the captured image data to the server, along with the user's identification information, and uploads the image data to the server via the Internet.

[0654] Input: Image data, user identification information

[0655] Output: None (data transfer to server)

[0656] Step 3:

[0657] The server analyzes the received image data and extracts handwriting features, such as line thickness, pen pressure, and curve angle, using an image processing library such as OpenCV, to extract numerical data.

[0658] Input: Image data

[0659] Output: Handwriting features (numerical data)

[0660] Step 4:

[0661] The server uses a machine learning algorithm to generate a handwriting model of the user based on the extracted handwriting features and stores it in a database. TensorFlow and PyTorch are used to learn the handwriting features and generate a model specific to the user.

[0662] Input: Handwriting features

[0663] Output: Handwritten model (model data)

[0664] Step 5:

[0665] When a user creates a customized product in the virtual store, the user inputs an input string (e.g., "custom message") into the terminal and transmits this data to the server. The terminal then transmits the text data entered by the user and the user identification information to the server.

[0666] Input: Text data (input string), user identification information

[0667] Output: None (data transfer to server)

[0668] Step 6:

[0669] The server searches for and retrieves the corresponding handwriting model from the database based on the user identification information, searches for the saved handwriting model, and loads the corresponding model into memory.

[0670] Input: User Identification Information

[0671] Output: Handwritten model (model data)

[0672] Step 7:

[0673] The server uses a handwriting model to generate an image that resembles the user's handwriting. A generative adversarial network (GAN) is used to convert the input text into a handwriting-imitating image.

[0674] Input: Text data, handwriting model

[0675] Output: Handwriting imitation image

[0676] Step 8:

[0677] The server overlays the generated handwriting imitation image onto the customized product image. Using PIL, the handwriting imitation image is synthesized onto the customized product design image.

[0678] Input: Handwriting imitation image, customized product image

[0679] Output: Customized product image (after compositing)

[0680] Step 9:

[0681] The server transmits the synthesized customized product image to the user terminal, and the generated customized product image is transmitted to the user terminal via the Internet and displayed.

[0682] Input: Customized product image (after compositing)

[0683] Output: None (data transfer to user terminal)

[0684] Step 10:

[0685] The user terminal displays the received customized product image to the user, and the image is displayed on the terminal display so that the user can check the product customized with their own handwriting.

[0686] Input: Customized product image (after receiving)

[0687] Output: User confirms customized product image

[0688] 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.

[0689] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[0690] 1. Acquisition of handwriting data and emotion data

[0691] Program processing:

[0692] The user writes characters by hand in an input field on the device. The device captures the handwritten characters as image data. The device also simultaneously captures the user's voice and facial expressions while writing using a camera and microphone. The acquired voice and facial expression data are sent to an emotion engine, which analyzes the user's emotional state. The image data and emotional state information are then sent to the server.

[0693] Examples:

[0694] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[0695] 2. Learning handwriting and emotion models

[0696] Program processing:

[0697] The server analyzes the received image data and extracts handwriting features, such as pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate a handwriting model and emotion model specific to the user and stores them in a database.

[0698] Examples:

[0699] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[0700] 3. Entering text and emotions

[0701] Program processing:

[0702] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[0703] Examples:

[0704] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[0705] 4. Generating handwriting imitation images

[0706] Program processing:

[0707] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[0708] Examples:

[0709] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[0710] 5. Displaying the generated results

[0711] Program processing:

[0712] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0713] Examples:

[0714] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[0715] This invention allows us to faithfully reproduce the handwriting and emotional state of individual users in a digital environment, thereby adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, etc.

[0716] The processing flow will be explained below.

[0717] Step 1:

[0718] The user writes characters by hand in an input field on the terminal.

[0719] The device captures the handwritten characters as image data.

[0720] Step 2:

[0721] The terminal transmits the captured handwritten character image and user identification information to the server.

[0722] Step 3:

[0723] The device captures the user's voice and facial expressions through a camera and microphone and sends them to the emotion engine, which analyzes the user's emotional state and sends the results to the server.

[0724] Step 4:

[0725] The server analyzes the received image data and extracts handwriting characteristics, such as line thickness, pressure, angle, and stroke order.

[0726] Step 5:

[0727] The server generates an emotion model based on the emotion state information received from the emotion engine, and the emotion model includes adjustment parameters for handwriting features according to the user's emotion state.

[0728] Step 6:

[0729] Based on the extracted handwriting features and emotion model, the server uses a machine learning algorithm to generate a handwriting model specific to the user and stores it in a database.

[0730] Step 7:

[0731] The user inputs a desired character string on the terminal.

[0732] The terminal transmits the character string, the user identification information, and the emotional state to the server.

[0733] Step 8:

[0734] The server retrieves a corresponding handwriting model from the database based on the user identification information, and further retrieves an emotion model based on the emotion state information.

[0735] Step 9:

[0736] The server uses the acquired handwriting model and emotion model to generate a handwriting imitation image of the input string that resembles the user's own handwriting. The style and angle of the characters are adjusted depending on the user's emotional state.

[0737] Step 10:

[0738] The server transmits the generated handwriting imitation image to the terminal.

[0739] Step 11:

[0740] The terminal displays the received handwriting imitation image on the user's screen.

[0741] For example, when a user writes "good morning" on a device, the device captures the characters as image data and sends it to the server. If the server recognizes that the user is in a "joy" emotional state, it performs analysis based on this information. When the user then types "hello," the server generates a handwriting imitation image with slightly thicker lines and softer curves, in line with the "joy" emotional state, and displays it on the device.

[0742] This invention makes it possible to generate character images that reflect the handwriting and emotions of individual users even in a digital environment, thereby imparting individuality and emotion to digital text. This opens up a wide range of applications, including handwritten-style notes, letters, and electronic signatures.

[0743] Example 2

[0744] 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."

[0745] Conventional systems can reproduce the characteristics of a user's handwriting when digitizing the user's handwritten text, but they are unable to reflect the user's emotional state. This makes it difficult to impart the user's personality and emotions to the digital text, resulting in a loss of the warmth and emotional expression inherent in handwritten text. The present invention aims to solve this problem by faithfully reproducing the user's handwriting and emotions in a digital environment.

[0746] 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.

[0747] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for analyzing the extracted handwriting features and emotional state to generate and save a handwriting model and emotional model dedicated to the user, means for receiving a character string input by the user and the emotional state and using the saved handwriting model and emotional model to generate an image that resembles the input character string and reflects the user's emotional state, and means for transmitting the generated image to a user terminal and displaying it. This makes it possible to generate a handwriting imitation image that reflects the user's individuality and emotions even in a digital environment.

[0748] The "means for acquiring handwritten character samples input by a user" refers to a device or method for capturing handwritten characters input by a user using a touch screen or a stylus pen and acquiring them as digital data.

[0749] The "means for extracting handwriting characteristics" refers to a device or method that analyzes and extracts characteristics such as writing pressure, line thickness, angle, and curve from an acquired handwritten character sample.

[0750] "Means for capturing emotional data" refers to a device or method that uses a camera or microphone to capture the user's facial expressions and voice and obtain them as digital data.

[0751] "Means for analyzing emotional state" refers to a device or method that analyzes captured facial expression and voice data to determine the emotional state of the user.

[0752] "Means for generating and storing handwriting models and emotion models" refers to a device or method that performs machine learning based on extracted handwriting features and analyzed emotional states, generates handwriting models and emotion models specific to the user, and stores them in a database.

[0753] The "means for generating an image that resembles the user's handwriting from an input character string and reflects the user's emotional state" refers to a device or method for generating a handwriting imitation image that resembles the user's handwriting, based on the character string and emotional state input by the user, using a stored handwriting model and emotional model.

[0754] The "means for transmitting the generated image to the user terminal and displaying it" refers to a device or method for transmitting the generated handwriting imitation image to the user terminal and displaying it on the screen of the terminal.

[0755] MODE FOR CARRYING OUT THE INVENTION

[0756] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[0757] Acquiring handwriting and emotion data

[0758] The user writes characters by hand in the input field on the device. The device captures this handwritten character as image data. Specifically, the device captures the user writing "good morning" on the touch screen with a stylus pen or finger and saves it as image data. The device also simultaneously captures the user's facial expressions and voice as they write using a camera and microphone. The acquired voice and facial expression data is sent to an emotion engine, which analyzes the user's emotional state. The device then sends this data to a server.

[0759] Examples:

[0760] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[0761] Learning handwriting and emotion models

[0762] The server analyzes the received image data and extracts handwriting features, specifically pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate handwriting and emotion models specific to the user and stores them in a database. Machine learning libraries used include TensorFlow and PyTorch.

[0763] Examples:

[0764] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[0765] String and emotion input

[0766] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[0767] Examples:

[0768] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[0769] Generating handwriting imitation images

[0770] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[0771] Examples:

[0772] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[0773] Viewing the generated results

[0774] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0775] Examples:

[0776] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[0777] Example prompt sentence:

[0778] A user writes "Good morning" in an input field by hand, and their voice (cheerful voice) and facial expression (happy smile) are captured to analyze their emotional state. Next, the same user types "Hello" and selects "Happy" as their emotional state. Using this information, the server should generate a handwriting imitation image of "Hello."

[0779] By using these steps, we can faithfully reproduce the user's handwriting and emotional state in a digital environment, adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, and more.

[0780] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0781] Step 1: User inputs a sample of their own handwriting

[0782] The user writes characters by hand using a stylus pen or their finger in the input field of the device. The device acquires this input and saves it as digital image data. A specific input is a string of characters such as "Good morning." The output is image data of the string of characters.

[0783] Step 2: Capture handwriting and emotion data

[0784] The device uses a camera and microphone to capture the user's facial expressions and voice while they are writing. The camera and microphone signals are processed to obtain facial expression data and voice data. Specifically, the device captures the user's smile and cheerful voice while they are writing "Good morning." The output is facial expression data and voice data.

[0785] Step 3: Analyze emotional state

[0786] The device sends the captured facial expression and voice data to the emotion engine, which analyzes the data and determines the user's emotional state. The specific inputs are facial expression and voice data, and the output is emotional state information such as "joy."

[0787] Step 4: Sending data

[0788] The terminal transmits image data of the handwritten characters and the analyzed emotional state to the server. The specific inputs are the image data and the emotional state information, and the output is the data arriving at the server.

[0789] Step 5: Extracting handwriting features

[0790] The server analyzes the received image data and extracts handwriting features such as pressure, line thickness, angle, and curve. Specifically, it uses an image processing algorithm to extract these features from the input data (handwriting image). The output is the extracted handwriting feature data.

[0791] Step 6: Generate emotion and handwriting models

[0792] The server applies machine learning algorithms to the extracted handwriting features and emotional state information to generate a user-specific handwriting model and emotion model. Specifically, it uses machine learning libraries such as TensorFlow and PyTorch. The input is the handwriting feature data and emotional state information, and the output is the generated model.

[0793] Step 7: Save the model

[0794] The server stores the generated handwriting and emotion models in a database. Specifically, it serializes the models and writes them to the database. The input is the model data, and the output is the results stored in the database.

[0795] Step 8: Entering Strings and Emotional States

[0796] The user inputs a desired string (e.g., "hello") and emotional state on the terminal. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The input is the string and emotional state information. The output is the input data on the terminal.

[0797] Step 9: Sending data (input text and emotional state)

[0798] The terminal sends the input string and emotional state to the server. The input is the string and emotional state information, and the output is the data arriving at the server.

[0799] Step 10: Getting the Model

[0800] The server retrieves the corresponding handwriting model and emotion model from the database. The input is the user identification information, and the output is the model data.

[0801] Step 11: Generate handwriting imitation images

[0802] The server uses the acquired model to generate a handwriting imitation image that matches the user's handwriting and emotional state. Specifically, it uses a generative adversarial network (GAN). The input is the string of characters, emotional state information, and model data, and the output is a handwriting imitation image.

[0803] Step 12: Sending the generated results

[0804] The server sends the generated handwriting imitation image to the terminal. The input is the handwriting imitation image, and the output is the image data arriving at the terminal.

[0805] Step 13: Display on the screen

[0806] The terminal displays the received handwriting imitation image on the user's screen. The input is the handwriting imitation image data, and the output is the display result on the user's screen.

[0807] (Application example 2)

[0808] 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."

[0809] Conventional systems generate handwriting imitation images based only on a user's handwriting samples, making it impossible to reflect changes in emotions or mood. Furthermore, there was a lack of a way to visually convey individual customers' emotions and personalities when improving customer experiences and promoting them in physical stores. In particular, there were limited ways for customers to share their emotions and feedback with other customers in the store. This made it difficult to provide an engaging customer experience using digital signage and displays.

[0810] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0811] In this invention, the server includes means for acquiring handwritten character samples and emotional states input by a user, means for analyzing the acquired handwritten character samples and emotional states to extract handwriting features and emotional features, and means for learning and saving a handwriting model and emotional model of the user based on the extracted handwriting features and emotional features, thereby enabling the generation of handwriting imitation images that reflect the user's individuality and emotions.

[0812] A "handwritten character sample" is a sample of characters input by handwriting by the user.

[0813] "Emotional state" is information that indicates the user's emotional or psychological state.

[0814] "Handwriting characteristics" refers to characteristics such as the shape, writing pressure, line thickness, angle, and curve of the user's handwritten characters.

[0815] "Emotional features" refer to features such as facial expressions, voice, and writing style based on the user's emotional state.

[0816] A "handwriting model" is a model that reproduces a user's unique handwriting, generated by a machine learning algorithm based on a user's handwriting samples.

[0817] The "emotion model" is a model for analyzing the user's emotional state and generating a handwriting imitation image that reflects the emotion.

[0818] "Image input means" refers to means for obtaining handwritten character samples using a camera, scanner, or the like.

[0819] The "emotion analysis means" is a means for analyzing voice and facial expression data to identify the emotional state of the user.

[0820] The "generating means" is a means for converting a character string input by a user into a handwriting imitation image using the stored handwriting model and emotion model.

[0821] The "database" is a storage device that stores the generated handwriting model and emotion model and makes them searchable as needed.

[0822] "User identification information" is information for uniquely identifying a user.

[0823] A "handwriting imitation image" is an image that reflects the user's handwriting and emotions, generated using a handwriting model and emotion model of the user.

[0824] Specific embodiments for carrying out the present invention will be described below.

[0825] The present invention is a system that generates handwritten-style message images based on a customer's handwriting sample and emotional state and displays them on an in-store display to improve customer experience in brick-and-mortar stores. This system generates a handwriting model and an emotion model based on user input data and uses them to generate a handwriting imitation image.

[0826] Hardware Configuration

[0827] Input device: Use an iPad or Android tablet, which allows customers to input handwritten messages.

[0828] Camera: Uses the device's built-in camera, which allows capturing customer facial expressions and analyzing their emotional state.

[0829] Microphone: Uses the device's built-in microphone, which allows us to capture the customer's voice and analyze their emotional state.

[0830] Server: Use a cloud server (e.g., AWS, Google Cloud) to process large amounts of data and generate models.

[0831] Display: A large display will be used in the store to display the generated handwritten message image.

[0832] Software Configuration

[0833] Handwriting capture: The device app combines a journal and note-taking app with the ability to accept handwriting input from customers.

[0834] Emotion analysis engine: Uses software (e.g., FaceAPI, Azure Cognitive Services) to capture voice and facial expression data and analyze emotions.

[0835] Data transmission: It has network functionality to transmit captured data to a cloud server in real time.

[0836] Model generation: On a cloud server, a handwriting model and an emotion model of the user are generated using a generative adversarial network (GAN) or a machine learning algorithm (e.g., TensorFlow, PyTorch).

[0837] Image generation & display: The generated handwriting imitation image is sent to the display via a web service and displayed in real time.

[0838] Specific examples

[0839] 1. Customer operations:

[0840] He writes in his own handwriting on a tablet inside the store, "I love this store!"

[0841] Cameras and microphones capture customers' facial expressions and voices.

[0842] 2. Emotion analysis:

[0843] The emotion analysis engine analyzes facial expressions and voice data to detect the emotional state of "joy."

[0844] 3. Data transmission:

[0845] The handwritten character sample and emotion data are sent to a cloud server.

[0846] 4. Model generation and image output:

[0847] The server analyzes handwriting features and emotion features based on the received data and generates a handwriting model and emotion model.

[0848] The sentence "I love this store!" is converted into a handwritten style that matches the emotion of joy, and is generated in slightly thicker strokes.

[0849] 5. Results display:

[0850] The generated handwriting imitation image is displayed on a large display inside the store.

[0851] The display features a hand-drawn image that reflects the customer's joyful emotions and reads, "We love this store!", for all customers to see.

[0852] Example prompts for generative AI models

[0853] "Generate a handwritten message image based on the handwritten characters and emotion data below.

[0854] Text: I love this store!

[0855] Emotion: Joy

[0856] Format: Handwritten, light background, thick strokes

[0857] This system will improve the customer experience in physical stores and make them feel more connected to other customers.

[0858] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0859] Step 1:

[0860] The device captures the user's handwritten text in the input field as image data, while the device's built-in camera and microphone simultaneously capture the user's facial expressions and voice, providing data for emotion analysis.

[0861] Input: Handwritten text, facial expression data, voice data

[0862] Output: Image data (handwritten characters), emotion analysis data (facial expressions, voice)

[0863] Step 2:

[0864] The device transmits the captured image data and sentiment analysis data to a server, which applies an image analysis algorithm to extract handwriting features from the image data and uses a sentiment analysis engine to extract sentiment features from the sentiment analysis data.

[0865] Input: Image data, emotion analysis data

[0866] Output: handwriting features, emotion features

[0867] Step 3:

[0868] The server uses the extracted handwriting features and emotion features to generate a user-specific handwriting model and emotion model using a machine learning algorithm (e.g., TensorFlow, PyTorch), and stores this in a database.

[0869] Input: handwriting features, emotion features

[0870] Output: Handwriting model, emotion model

[0871] Step 4:

[0872] The user inputs a desired character string and their emotional state on the terminal. The terminal transmits the data to the server. The server uses the stored handwriting model and emotion model to generate a handwriting imitation image that resembles the user's handwriting and corresponds to their emotional state.

[0873] Input: Input string, emotional state

[0874] Output: Handwriting imitation image

[0875] Step 5:

[0876] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen and also sends it to a large display in the store to share with other customers.

[0877] Input: Handwriting imitation image

[0878] Output: Images displayed on user devices and large displays

[0879] 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.

[0880] 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.

[0881] 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.

[0882] [Third embodiment]

[0883] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0884] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0885] 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).

[0886] 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.

[0887] 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.

[0888] 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).

[0889] 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. 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.

[0890] 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.

[0891] 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.

[0892] 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.

[0893] 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.

[0894] 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."

[0895] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs a character string input by the user as an image that resembles the user's handwriting. A specific embodiment of the system will be described below.

[0896] 1. Acquiring handwriting data

[0897] Program processing:

[0898] A user writes characters by hand in an input field provided on a device (smartphone, tablet, etc.). The device captures the handwritten characters as image data and sends it to a server. This image data also includes the user's identification information.

[0899] Examples:

[0900] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0901] 2. Training the handwriting model

[0902] Program processing:

[0903] The server analyzes the received image data and extracts handwriting features, such as line thickness, pressure, curve angle, and stroke order. Based on the extracted handwriting features, a machine learning algorithm is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0904] Examples:

[0905] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0906] 3. Entering a string

[0907] Program processing:

[0908] The user inputs a desired character string into the terminal, which then transmits the input character string and the user's identification information to the server.

[0909] Examples:

[0910] The user enters the string "Hello" as text into the input field of the terminal, which then sends it to the server.

[0911] 4. Generating handwriting imitation images

[0912] Program processing:

[0913] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are often used to generate the image.

[0914] Examples:

[0915] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0916] 5. Displaying the generated results

[0917] Program processing:

[0918] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[0919] Examples:

[0920] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0921] This system can faithfully reproduce the handwriting of individual users in a digital environment, adding individuality to digital text. This system can be used for a variety of purposes, including handwritten notes, letters, and electronic signatures.

[0922] The processing flow will be explained below.

[0923] Step 1:

[0924] The user writes characters by hand in an input field on the terminal.

[0925] The device captures the handwritten characters as image data.

[0926] Step 2:

[0927] The terminal transmits the captured handwritten character image and user identification information to the server.

[0928] Step 3:

[0929] The server analyzes the received image data and extracts handwriting features.

[0930] Specifically, it analyzes characteristics such as line thickness, pressure, angle, and stroke order.

[0931] Step 4:

[0932] The server uses a machine learning algorithm to generate a handwriting model specific to the user based on the extracted handwriting features.

[0933] Step 5:

[0934] The server stores the generated handwriting model in a database.

[0935] Step 6:

[0936] The user inputs a desired character string into an input field on the terminal.

[0937] Step 7:

[0938] The terminal transmits the input string and the user identification information to the server.

[0939] Step 8:

[0940] The server retrieves the corresponding handwriting model from the database based on the user identification information.

[0941] Step 9:

[0942] The server uses the acquired handwriting model to generate a handwriting imitation image that resembles the user's own handwriting for the input character string.

[0943] Step 10:

[0944] The server transmits the generated handwriting imitation image to the terminal.

[0945] Step 11:

[0946] The terminal displays the received handwriting imitation image on the user's screen.

[0947] Example 1

[0948] 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."

[0949] Conventional digital text generation methods have difficulty faithfully reproducing the handwriting of individual users, making it impossible to create unique digital content. Therefore, there is a demand for technology that can accurately reproduce a user's handwriting even in a digital environment and generate unique handwritten-style notes, letters, electronic signatures, etc.

[0950] 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.

[0951] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string that resembles the user's handwriting using the saved handwriting model, and means for transmitting the generated image to a user terminal and displaying it, thereby enabling the generation of digital text that accurately reproduces the user's handwriting.

[0952] "User input" refers to manual operations or data input by a user into the system.

[0953] A "handwritten character sample" is a digital image of a character handwritten by a user.

[0954] "Analysis" is the process of analyzing data and information in detail and extracting its meaning and characteristics.

[0955] "Handwriting features" are characteristic information about a particular handwriting, such as the thickness of the lines of handwritten characters, the pressure of the writing, the angle of the curves, and the order of strokes.

[0956] A "handwriting model" is a machine learning model that learns the user's unique handwriting characteristics.

[0957] "Learning" is the process by which machine learning algorithms understand patterns and features based on data.

[0958] "Storage" is the act of permanently recording data or information on a storage medium.

[0959] "Generating an image" means creating new image data using a specific algorithm.

[0960] A "user terminal" is a device (e.g., a smartphone, tablet, etc.) that a user uses to access the system.

[0961] "Display" refers to the act of visually presenting generated data or information to a user.

[0962] "Identification information" is data for uniquely identifying a user (for example, a user ID or authentication information).

[0963] "Image input means" refers to a device or function that captures a user's handwritten characters as a digital image.

[0964] "Image analysis" is the process of extracting information from digital images and analyzing it.

[0965] A "machine learning algorithm" is an algorithm that improves itself based on data.

[0966] A "database" is a system for efficiently managing, storing, and retrieving large amounts of data.

[0967] A "handwriting imitation image" is a digital image generated to resemble a user's own handwriting.

[0968] "Generation means" refers to the functions or devices that allow the system to create or convert new data or information.

[0969] A "prompt sentence" is text data that is input into an AI model.

[0970] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs an image of a character string input by the user that resembles the user's handwriting. Specific embodiments of the system are described below.

[0971] Acquiring handwriting data

[0972] A user writes characters by hand in an input field provided on a device such as a smartphone or tablet. When the user writes characters using a stylus pen or their finger, the device captures the characters as image data. The device then sends this image data and the user's identification information to a server. The hardware used is a typical smartphone or tablet, and the software has an image capture function built in.

[0973] Examples:

[0974] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[0975] Training a handwriting model

[0976] The server analyzes the received image data and extracts handwriting features. This analysis uses an image processing library (e.g., OpenCV). Specifically, features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. Based on the extracted handwriting features, a machine learning algorithm (e.g., TensorFlow or PyTorch) is used to generate a handwriting model specific to the user, and this model is stored in a database.

[0977] Examples:

[0978] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[0979] Entering a string

[0980] The user enters a desired string on the device. For example, they enter "hello" into a text input field. The device then sends the string and user identification information to the server. The hardware used includes a typical smartphone or tablet input field.

[0981] Examples:

[0982] The user enters the string "hello" as text into the terminal, which then sends it to the server.

[0983] Generating handwriting imitation images

[0984] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image.

[0985] Examples:

[0986] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[0987] Viewing the generated results

[0988] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen, allowing the user to confirm that the input string appears similar to their own handwriting.

[0989] Examples:

[0990] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[0991] Example prompt sentence:

[0992] "I would like to develop a system that acquires a user's handwriting, analyzes it using a machine learning model, and generates a handwritten-like image. Please tell me the specific steps: the user inputs characters into the device → the server analyzes the handwriting and generates a model → the input string is converted into a handwritten-like image and displayed. Please also tell me the technology and algorithms used."

[0993] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0994] Step 1:

[0995] The user writes handwritten characters on the device.

[0996] The user writes handwritten characters in an input field provided on the screen of a smartphone or tablet. The user inputs using a stylus pen or a finger. The device captures the handwritten characters as image data. The input data is an image of the handwritten characters, and the output data is the captured image file.

[0997] Specific behavior:

[0998] The user writes the words "Good morning" in the input field.

[0999] The device captures this text as an image and saves it in an internal temporary file.

[1000] Step 2:

[1001] The device captures the handwritten image and sends it to the server.

[1002] The device sends the captured image data and the user's identification information to the server. The input data is the image and user identification information acquired in the previous step, and the output data is the HTTP request sent to the server.

[1003] Specific behavior:

[1004] The device sends the captured image data of "Good morning" to the server as an HTTP POST request.

[1005] Step 3:

[1006] The server analyzes the image data and extracts handwriting features.

[1007] The server analyzes the received image data and uses an image processing library (e.g., OpenCV) to extract handwriting features. Features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. The input data is an image file, and the output data is the extracted handwriting feature data.

[1008] Specific behavior:

[1009] The server reads the "Good morning" image file and converts it to grayscale.

[1010] The server quantifies the thickness of the lines and the angle of the curves and extracts this as feature data.

[1011] Step 4:

[1012] The server uses machine learning algorithms to generate and store handwriting models.

[1013] The server uses the extracted handwriting feature data to generate a user-specific handwriting model using a machine learning algorithm (e.g., TensorFlow or PyTorch). The generated model is stored in a database. The input data is the handwriting feature data, and the output data is the trained handwriting model.

[1014] Specific behavior:

[1015] The server uses a neural network to learn the feature data and generate a handwriting model for the user.

[1016] Save the generated model in the database.

[1017] Step 5:

[1018] The user enters a string into a text input field on the device.

[1019] The user inputs a desired string in text format into the terminal. The input data is the text string and user identification information, and the output data is an HTTP request sent to the server.

[1020] Specific behavior:

[1021] The user enters the string "Hello" into the text input field on the device.

[1022] The terminal transmits this text data and the user identification information to the server.

[1023] Step 6:

[1024] The server generates a handwriting imitation image using the user's handwriting model.

[1025] The server retrieves the corresponding handwriting model from the database based on the user's identification information. Using the retrieved model, it generates an image that resembles the user's handwriting from the input string. GAN or other image generation algorithms are used for image generation. The input data are the text string and the handwriting model, and the output data is a handwriting imitation image.

[1026] Specific behavior:

[1027] The server reads the corresponding user's handwriting model from the database.

[1028] Using the text string "hello" as input, a handwriting imitation image is generated using the GAN algorithm.

[1029] Step 7:

[1030] The server generates a handwriting imitation image and sends it to the terminal, which displays it.

[1031] The server sends the generated handwriting imitation image to the terminal, which displays the received image on the user's screen. The input data is the handwriting imitation image, and the output data is the displayed image.

[1032] Specific behavior:

[1033] The server sends the generated handwriting imitation image of "hello" to the terminal as an HTTP response.

[1034] The terminal receives this image, places it on the display area, and presents it to the user.

[1035] (Application example 1)

[1036] 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."

[1037] In conventional virtual stores, it was difficult for users to input messages in their own handwriting and have them reflected in customized products. This made it difficult to provide custom-made products that reflected each user's individuality in a digital environment, posing challenges in improving user satisfaction. Furthermore, existing systems lack a method for faithfully reproducing a user's handwriting, and there is a lack of technology for creating handwritten-style customized products.

[1038] 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.

[1039] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string by using the saved handwriting model to resemble the user's handwriting, means for adding a message that looks like the user's handwriting to customized products in the virtual store, and means for transmitting the generated image to a user terminal and displaying it. This enables users to easily create and order customized products with original designs using their own handwriting.

[1040] "User input" includes characters and information entered by a user using a device.

[1041] A "handwriting sample" is a character written by a user in their own handwriting on a device.

[1042] "Handwriting characteristics" refer to characteristics such as the thickness of lines, writing pressure, and angle of curves that are specific to the user's handwritten characters.

[1043] A "handwriting model" is a digital model generated using a machine learning algorithm based on the user's handwriting characteristics.

[1044] "Image generation" is a process of generating an image that resembles the user's handwriting using a handwriting model based on input text.

[1045] A "virtual store" is a virtual store operated on the Internet where users can select and purchase products.

[1046] A "customized product" is a product that is individually created by a user to add specific elements (in this case, a handwritten message) to suit their preferences.

[1047] "User identification information" is information for identifying a specific user, and typically includes a user ID and authentication information.

[1048] The "image input means" is a means for capturing characters written by a user on a device as an image.

[1049] The "generation means" is a means for converting a character string input by a user into a handwriting imitation image using a specific algorithm.

[1050] The "display means" is a means for displaying the generated handwriting imitation image on the user terminal.

[1051] MODE FOR CARRYING OUT THE INVENTION

[1052] System Configuration

[1053] The system for implementing this invention consists of a series of processes: a user inputs handwritten characters on a terminal, a handwriting model is generated for the user based on the input, and finally, the user's handwritten-style message is added to a customized product in a virtual store. The specific system configuration is described below.

[1054] Hardware Configuration

[1055] User terminals include smartphones, tablets, PCs, etc. These terminals are equipped with image input means (cameras, touch screens, etc.).

[1056] The server used is a cloud server, which has the capacity and database to run machine learning algorithms.

[1057] Software Configuration

[1058] The overall system utilizes the following software:

[1059] 1. Image analysis software: Using image processing libraries such as OpenCV, we analyze the user's handwritten character samples and extract handwriting features.

[1060] 2. Machine learning algorithms: Using TensorFlow and PyTorch, we generate a handwriting model of the user based on the extracted handwriting features.

[1061] 3. Generative Adversarial Network (GAN): Using a GAN, an image of the input string is generated that resembles the user's handwriting based on a model of the user's own handwriting.

[1062] 4. Image editing software: Use PIL (Python Imaging Library) to overlay the generated handwriting imitation image onto the customized product image.

[1063] Program processing

[1064] The server captures handwriting samples entered by the user on the device and transmits them to the server. The captured image is analyzed on the server side to extract handwriting features. Based on the extracted features, a machine learning algorithm is used to generate a handwriting model of the user and store it in a database.

[1065] When a user wishes to create a customized product, a request is sent from the terminal along with a character string. The server searches for a handwriting model based on the user's identification information and converts the input character string into a handwriting imitation image. The generated handwriting imitation image is overlaid on the customized product using the PIL library, and finally sent to the user's terminal for display.

[1066] Specific examples

[1067] For example, when a user writes "thank you" by hand using a smartphone, the character is captured as an image and sent to the server. The server analyzes the image and extracts handwriting features. Based on this, a handwriting model of the user is generated and stored in a database.

[1068] When a user customizes a T-shirt in the virtual store, they input the word "thank you" and send the text data to the server. The server uses a model of the user's handwriting to convert the input text into a handwriting imitation image and overlay this image on the T-shirt design. Finally, the generated T-shirt design is displayed on the user's smartphone.

[1069] Prompt Sentence Examples

[1070] When a user makes a request to the server for a customized product, the prompt text can be written as follows:

[1071] User ID: 12345, Text: "Thank you"

[1072] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1073] Step 1:

[1074] The user inputs handwritten characters on the device. The user writes characters such as "Thank you" by hand on the touch screen of a smartphone or tablet. This input is saved as image data on the device.

[1075] Input: Handwritten characters

[1076] Output: Image data

[1077] Step 2:

[1078] The device sends the captured image data to the server, along with the user's identification information, and uploads the image data to the server via the Internet.

[1079] Input: Image data, user identification information

[1080] Output: None (data transfer to server)

[1081] Step 3:

[1082] The server analyzes the received image data and extracts handwriting features, such as line thickness, pen pressure, and curve angle, using an image processing library such as OpenCV, to extract numerical data.

[1083] Input: Image data

[1084] Output: Handwriting features (numerical data)

[1085] Step 4:

[1086] The server uses a machine learning algorithm to generate a handwriting model of the user based on the extracted handwriting features and stores it in a database. TensorFlow and PyTorch are used to learn the handwriting features and generate a model specific to the user.

[1087] Input: Handwriting features

[1088] Output: Handwritten model (model data)

[1089] Step 5:

[1090] When a user creates a customized product in the virtual store, the user inputs an input string (e.g., "custom message") into the terminal and transmits this data to the server. The terminal then transmits the text data entered by the user and the user identification information to the server.

[1091] Input: Text data (input string), user identification information

[1092] Output: None (data transfer to server)

[1093] Step 6:

[1094] The server searches for and retrieves the corresponding handwriting model from the database based on the user identification information, searches for the saved handwriting model, and loads the corresponding model into memory.

[1095] Input: User Identification Information

[1096] Output: Handwritten model (model data)

[1097] Step 7:

[1098] The server uses a handwriting model to generate an image that resembles the user's handwriting. A generative adversarial network (GAN) is used to convert the input text into a handwriting-imitating image.

[1099] Input: Text data, handwriting model

[1100] Output: Handwriting imitation image

[1101] Step 8:

[1102] The server overlays the generated handwriting imitation image onto the customized product image. Using PIL, the handwriting imitation image is synthesized onto the customized product design image.

[1103] Input: Handwriting imitation image, customized product image

[1104] Output: Customized product image (after compositing)

[1105] Step 9:

[1106] The server transmits the synthesized customized product image to the user terminal, and the generated customized product image is transmitted to the user terminal via the Internet and displayed.

[1107] Input: Customized product image (after compositing)

[1108] Output: None (data transfer to user terminal)

[1109] Step 10:

[1110] The user terminal displays the received customized product image to the user, and the image is displayed on the terminal display so that the user can check the product customized with their own handwriting.

[1111] Input: Customized product image (after receiving)

[1112] Output: User confirms customized product image

[1113] 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.

[1114] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[1115] 1. Acquisition of handwriting data and emotion data

[1116] Program processing:

[1117] The user writes characters by hand in an input field on the device. The device captures the handwritten characters as image data. The device also simultaneously captures the user's voice and facial expressions while writing using a camera and microphone. The acquired voice and facial expression data are sent to an emotion engine, which analyzes the user's emotional state. The image data and emotional state information are then sent to the server.

[1118] Examples:

[1119] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[1120] 2. Learning handwriting and emotion models

[1121] Program processing:

[1122] The server analyzes the received image data and extracts handwriting features, such as pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate a handwriting model and emotion model specific to the user and stores them in a database.

[1123] Examples:

[1124] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[1125] 3. Entering text and emotions

[1126] Program processing:

[1127] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[1128] Examples:

[1129] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[1130] 4. Generating handwriting imitation images

[1131] Program processing:

[1132] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[1133] Examples:

[1134] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[1135] 5. Displaying the generated results

[1136] Program processing:

[1137] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[1138] Examples:

[1139] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[1140] This invention allows us to faithfully reproduce the handwriting and emotional state of individual users in a digital environment, thereby adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, etc.

[1141] The processing flow will be explained below.

[1142] Step 1:

[1143] The user writes characters by hand in an input field on the terminal.

[1144] The device captures the handwritten characters as image data.

[1145] Step 2:

[1146] The terminal transmits the captured handwritten character image and user identification information to the server.

[1147] Step 3:

[1148] The device captures the user's voice and facial expressions through a camera and microphone and sends them to the emotion engine, which analyzes the user's emotional state and sends the results to the server.

[1149] Step 4:

[1150] The server analyzes the received image data and extracts handwriting characteristics, such as line thickness, pressure, angle, and stroke order.

[1151] Step 5:

[1152] The server generates an emotion model based on the emotion state information received from the emotion engine, and the emotion model includes adjustment parameters for handwriting features according to the user's emotion state.

[1153] Step 6:

[1154] Based on the extracted handwriting features and emotion model, the server uses a machine learning algorithm to generate a handwriting model specific to the user and stores it in a database.

[1155] Step 7:

[1156] The user inputs a desired character string on the terminal.

[1157] The terminal transmits the character string, the user identification information, and the emotional state to the server.

[1158] Step 8:

[1159] The server retrieves a corresponding handwriting model from the database based on the user identification information, and further retrieves an emotion model based on the emotion state information.

[1160] Step 9:

[1161] The server uses the acquired handwriting model and emotion model to generate a handwriting imitation image of the input string that resembles the user's own handwriting. The style and angle of the characters are adjusted depending on the user's emotional state.

[1162] Step 10:

[1163] The server transmits the generated handwriting imitation image to the terminal.

[1164] Step 11:

[1165] The terminal displays the received handwriting imitation image on the user's screen.

[1166] For example, when a user writes "good morning" on a device, the device captures the characters as image data and sends it to the server. If the server recognizes that the user is in a "joy" emotional state, it performs analysis based on this information. When the user then types "hello," the server generates a handwriting imitation image with slightly thicker lines and softer curves, in line with the "joy" emotional state, and displays it on the device.

[1167] This invention makes it possible to generate character images that reflect the handwriting and emotions of individual users even in a digital environment, thereby imparting individuality and emotion to digital text. This opens up a wide range of applications, including handwritten-style notes, letters, and electronic signatures.

[1168] Example 2

[1169] 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."

[1170] Conventional systems can reproduce the characteristics of a user's handwriting when digitizing the user's handwritten text, but they are unable to reflect the user's emotional state. This makes it difficult to impart the user's personality and emotions to the digital text, resulting in a loss of the warmth and emotional expression inherent in handwritten text. The present invention aims to solve this problem by faithfully reproducing the user's handwriting and emotions in a digital environment.

[1171] 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.

[1172] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for analyzing the extracted handwriting features and emotional state to generate and save a handwriting model and emotional model dedicated to the user, means for receiving a character string input by the user and the emotional state and using the saved handwriting model and emotional model to generate an image that resembles the input character string and reflects the user's emotional state, and means for transmitting the generated image to a user terminal and displaying it. This makes it possible to generate a handwriting imitation image that reflects the user's individuality and emotions even in a digital environment.

[1173] The "means for acquiring handwritten character samples input by a user" refers to a device or method for capturing handwritten characters input by a user using a touch screen or a stylus pen and acquiring them as digital data.

[1174] The "means for extracting handwriting characteristics" refers to a device or method that analyzes and extracts characteristics such as writing pressure, line thickness, angle, and curve from an acquired handwritten character sample.

[1175] "Means for capturing emotional data" refers to a device or method that uses a camera or microphone to capture the user's facial expressions and voice and obtain them as digital data.

[1176] "Means for analyzing emotional state" refers to a device or method that analyzes captured facial expression and voice data to determine the emotional state of the user.

[1177] "Means for generating and storing handwriting models and emotion models" refers to a device or method that performs machine learning based on extracted handwriting features and analyzed emotional states, generates handwriting models and emotion models specific to the user, and stores them in a database.

[1178] The "means for generating an image that resembles the user's handwriting from an input character string and reflects the user's emotional state" refers to a device or method for generating a handwriting imitation image that resembles the user's handwriting, based on the character string and emotional state input by the user, using a stored handwriting model and emotional model.

[1179] The "means for transmitting the generated image to the user terminal and displaying it" refers to a device or method for transmitting the generated handwriting imitation image to the user terminal and displaying it on the screen of the terminal.

[1180] MODE FOR CARRYING OUT THE INVENTION

[1181] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[1182] Acquiring handwriting and emotion data

[1183] The user writes characters by hand in the input field on the device. The device captures this handwritten character as image data. Specifically, the device captures the user writing "good morning" on the touch screen with a stylus pen or finger and saves it as image data. The device also simultaneously captures the user's facial expressions and voice as they write using a camera and microphone. The acquired voice and facial expression data is sent to an emotion engine, which analyzes the user's emotional state. The device then sends this data to a server.

[1184] Examples:

[1185] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[1186] Learning handwriting and emotion models

[1187] The server analyzes the received image data and extracts handwriting features, specifically pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate handwriting and emotion models specific to the user and stores them in a database. Machine learning libraries used include TensorFlow and PyTorch.

[1188] Examples:

[1189] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[1190] String and emotion input

[1191] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[1192] Examples:

[1193] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[1194] Generating handwriting imitation images

[1195] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[1196] Examples:

[1197] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[1198] Viewing the generated results

[1199] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[1200] Examples:

[1201] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[1202] Example prompt sentence:

[1203] A user writes "Good morning" in an input field by hand, and their voice (cheerful voice) and facial expression (happy smile) are captured to analyze their emotional state. Next, the same user types "Hello" and selects "Happy" as their emotional state. Using this information, the server should generate a handwriting imitation image of "Hello."

[1204] By using these steps, we can faithfully reproduce the user's handwriting and emotional state in a digital environment, adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, and more.

[1205] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1206] Step 1: User inputs a sample of their own handwriting

[1207] The user writes characters by hand using a stylus pen or their finger in the input field of the device. The device acquires this input and saves it as digital image data. A specific input is a string of characters such as "Good morning." The output is image data of the string of characters.

[1208] Step 2: Capture handwriting and emotion data

[1209] The device uses a camera and microphone to capture the user's facial expressions and voice while they are writing. The camera and microphone signals are processed to obtain facial expression data and voice data. Specifically, the device captures the user's smile and cheerful voice while they are writing "Good morning." The output is facial expression data and voice data.

[1210] Step 3: Analyze emotional state

[1211] The device sends the captured facial expression and voice data to the emotion engine, which analyzes the data and determines the user's emotional state. The specific inputs are facial expression and voice data, and the output is emotional state information such as "joy."

[1212] Step 4: Sending data

[1213] The terminal transmits image data of the handwritten characters and the analyzed emotional state to the server. The specific inputs are the image data and the emotional state information, and the output is the data arriving at the server.

[1214] Step 5: Extracting handwriting features

[1215] The server analyzes the received image data and extracts handwriting features such as pressure, line thickness, angle, and curve. Specifically, it uses an image processing algorithm to extract these features from the input data (handwriting image). The output is the extracted handwriting feature data.

[1216] Step 6: Generate emotion and handwriting models

[1217] The server applies machine learning algorithms to the extracted handwriting features and emotional state information to generate a user-specific handwriting model and emotion model. Specifically, it uses machine learning libraries such as TensorFlow and PyTorch. The input is the handwriting feature data and emotional state information, and the output is the generated model.

[1218] Step 7: Save the model

[1219] The server stores the generated handwriting and emotion models in a database. Specifically, it serializes the models and writes them to the database. The input is the model data, and the output is the results stored in the database.

[1220] Step 8: Entering Strings and Emotional States

[1221] The user inputs a desired string (e.g., "hello") and emotional state on the terminal. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The input is the string and emotional state information. The output is the input data on the terminal.

[1222] Step 9: Sending data (input text and emotional state)

[1223] The terminal sends the input string and emotional state to the server. The input is the string and emotional state information, and the output is the data arriving at the server.

[1224] Step 10: Getting the Model

[1225] The server retrieves the corresponding handwriting model and emotion model from the database. The input is the user identification information, and the output is the model data.

[1226] Step 11: Generate handwriting imitation images

[1227] The server uses the acquired model to generate a handwriting imitation image that matches the user's handwriting and emotional state. Specifically, it uses a generative adversarial network (GAN). The input is the string of characters, emotional state information, and model data, and the output is a handwriting imitation image.

[1228] Step 12: Sending the generated results

[1229] The server sends the generated handwriting imitation image to the terminal. The input is the handwriting imitation image, and the output is the image data arriving at the terminal.

[1230] Step 13: Display on the screen

[1231] The terminal displays the received handwriting imitation image on the user's screen. The input is the handwriting imitation image data, and the output is the display result on the user's screen.

[1232] (Application example 2)

[1233] 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."

[1234] Conventional systems generate handwriting imitation images based only on a user's handwriting samples, making it impossible to reflect changes in emotions or mood. Furthermore, there was a lack of a way to visually convey individual customers' emotions and personalities when improving customer experiences and promoting them in physical stores. In particular, there were limited ways for customers to share their emotions and feedback with other customers in the store. This made it difficult to provide an engaging customer experience using digital signage and displays.

[1235] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1236] In this invention, the server includes means for acquiring handwritten character samples and emotional states input by a user, means for analyzing the acquired handwritten character samples and emotional states to extract handwriting features and emotional features, and means for learning and saving a handwriting model and emotional model of the user based on the extracted handwriting features and emotional features, thereby enabling the generation of handwriting imitation images that reflect the user's individuality and emotions.

[1237] A "handwritten character sample" is a sample of characters input by handwriting by the user.

[1238] "Emotional state" is information that indicates the user's emotional or psychological state.

[1239] "Handwriting characteristics" refers to characteristics such as the shape, writing pressure, line thickness, angle, and curve of the user's handwritten characters.

[1240] "Emotional features" refer to features such as facial expressions, voice, and writing style based on the user's emotional state.

[1241] A "handwriting model" is a model that reproduces a user's unique handwriting, generated by a machine learning algorithm based on a user's handwriting samples.

[1242] The "emotion model" is a model for analyzing the user's emotional state and generating a handwriting imitation image that reflects the emotion.

[1243] "Image input means" refers to means for obtaining handwritten character samples using a camera, scanner, or the like.

[1244] The "emotion analysis means" is a means for analyzing voice and facial expression data to identify the emotional state of the user.

[1245] The "generating means" is a means for converting a character string input by a user into a handwriting imitation image using the stored handwriting model and emotion model.

[1246] The "database" is a storage device that stores the generated handwriting model and emotion model and makes them searchable as needed.

[1247] "User identification information" is information for uniquely identifying a user.

[1248] A "handwriting imitation image" is an image that reflects the user's handwriting and emotions, generated using a handwriting model and emotion model of the user.

[1249] Specific embodiments for carrying out the present invention will be described below.

[1250] The present invention is a system that generates handwritten-style message images based on a customer's handwriting sample and emotional state and displays them on an in-store display to improve customer experience in brick-and-mortar stores. This system generates a handwriting model and an emotion model based on user input data and uses them to generate a handwriting imitation image.

[1251] Hardware Configuration

[1252] Input device: Use an iPad or Android tablet, which allows customers to input handwritten messages.

[1253] Camera: Uses the device's built-in camera, which allows capturing customer facial expressions and analyzing their emotional state.

[1254] Microphone: Uses the device's built-in microphone, which allows us to capture the customer's voice and analyze their emotional state.

[1255] Server: Use a cloud server (e.g., AWS, Google Cloud) to process large amounts of data and generate models.

[1256] Display: A large display will be used in the store to display the generated handwritten message image.

[1257] Software Configuration

[1258] Handwriting capture: The device app combines a journal and note-taking app with the ability to accept handwriting input from customers.

[1259] Emotion analysis engine: Uses software (e.g., FaceAPI, Azure Cognitive Services) to capture voice and facial expression data and analyze emotions.

[1260] Data transmission: It has network functionality to transmit captured data to a cloud server in real time.

[1261] Model generation: On a cloud server, a handwriting model and an emotion model of the user are generated using a generative adversarial network (GAN) or a machine learning algorithm (e.g., TensorFlow, PyTorch).

[1262] Image generation & display: The generated handwriting imitation image is sent to the display via a web service and displayed in real time.

[1263] Specific examples

[1264] 1. Customer operations:

[1265] He writes in his own handwriting on a tablet inside the store, "I love this store!"

[1266] Cameras and microphones capture customers' facial expressions and voices.

[1267] 2. Emotion analysis:

[1268] The emotion analysis engine analyzes facial expressions and voice data to detect the emotional state of "joy."

[1269] 3. Data transmission:

[1270] The handwritten character sample and emotion data are sent to a cloud server.

[1271] 4. Model generation and image output:

[1272] The server analyzes handwriting features and emotion features based on the received data and generates a handwriting model and emotion model.

[1273] The sentence "I love this store!" is converted into a handwritten style that matches the emotion of joy, and is generated in slightly thicker strokes.

[1274] 5. Results display:

[1275] The generated handwriting imitation image is displayed on a large display inside the store.

[1276] The display features a hand-drawn image that reflects the customer's joyful emotions and reads, "We love this store!", for all customers to see.

[1277] Example prompts for generative AI models

[1278] "Generate a handwritten message image based on the handwritten characters and emotion data below.

[1279] Text: I love this store!

[1280] Emotion: Joy

[1281] Format: Handwritten, light background, thick strokes

[1282] This system will improve the customer experience in physical stores and make them feel more connected to other customers.

[1283] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1284] Step 1:

[1285] The device captures the user's handwritten text in the input field as image data, while the device's built-in camera and microphone simultaneously capture the user's facial expressions and voice, providing data for emotion analysis.

[1286] Input: Handwritten text, facial expression data, voice data

[1287] Output: Image data (handwritten characters), emotion analysis data (facial expressions, voice)

[1288] Step 2:

[1289] The device transmits the captured image data and sentiment analysis data to a server, which applies an image analysis algorithm to extract handwriting features from the image data and uses a sentiment analysis engine to extract sentiment features from the sentiment analysis data.

[1290] Input: Image data, emotion analysis data

[1291] Output: handwriting features, emotion features

[1292] Step 3:

[1293] The server uses the extracted handwriting features and emotion features to generate a user-specific handwriting model and emotion model using a machine learning algorithm (e.g., TensorFlow, PyTorch), and stores this in a database.

[1294] Input: handwriting features, emotion features

[1295] Output: Handwriting model, emotion model

[1296] Step 4:

[1297] The user inputs a desired character string and their emotional state on the terminal. The terminal transmits the data to the server. The server uses the stored handwriting model and emotion model to generate a handwriting imitation image that resembles the user's handwriting and corresponds to their emotional state.

[1298] Input: Input string, emotional state

[1299] Output: Handwriting imitation image

[1300] Step 5:

[1301] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen and also sends it to a large display in the store to share with other customers.

[1302] Input: Handwriting imitation image

[1303] Output: Images displayed on user devices and large displays

[1304] 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.

[1305] 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.

[1306] 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.

[1307] [Fourth embodiment]

[1308] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1309] 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.

[1310] 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).

[1311] 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.

[1312] 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.

[1313] 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).

[1314] 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. 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.

[1315] 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.

[1316] 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.

[1317] 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.

[1318] 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.

[1319] 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.

[1320] 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."

[1321] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs a character string input by the user as an image that resembles the user's handwriting. A specific embodiment of the system will be described below.

[1322] 1. Acquiring handwriting data

[1323] Program processing:

[1324] A user writes characters by hand in an input field provided on a device (smartphone, tablet, etc.). The device captures the handwritten characters as image data and sends it to a server. This image data also includes the user's identification information.

[1325] Examples:

[1326] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[1327] 2. Training the handwriting model

[1328] Program processing:

[1329] The server analyzes the received image data and extracts handwriting features, such as line thickness, pressure, curve angle, and stroke order. Based on the extracted handwriting features, a machine learning algorithm is used to generate a handwriting model specific to the user, and this model is stored in a database.

[1330] Examples:

[1331] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[1332] 3. Entering a string

[1333] Program processing:

[1334] The user inputs a desired character string into the terminal, which then transmits the input character string and the user's identification information to the server.

[1335] Examples:

[1336] The user enters the string "Hello" as text into the input field of the terminal, which then sends it to the server.

[1337] 4. Generating handwriting imitation images

[1338] Program processing:

[1339] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are often used to generate the image.

[1340] Examples:

[1341] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[1342] 5. Displaying the generated results

[1343] Program processing:

[1344] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[1345] Examples:

[1346] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[1347] This system can faithfully reproduce the handwriting of individual users in a digital environment, adding individuality to digital text. This system can be used for a variety of purposes, including handwritten notes, letters, and electronic signatures.

[1348] The processing flow will be explained below.

[1349] Step 1:

[1350] The user writes characters by hand in an input field on the terminal.

[1351] The device captures the handwritten characters as image data.

[1352] Step 2:

[1353] The terminal transmits the captured handwritten character image and user identification information to the server.

[1354] Step 3:

[1355] The server analyzes the received image data and extracts handwriting features.

[1356] Specifically, it analyzes characteristics such as line thickness, pressure, angle, and stroke order.

[1357] Step 4:

[1358] The server uses a machine learning algorithm to generate a handwriting model specific to the user based on the extracted handwriting features.

[1359] Step 5:

[1360] The server stores the generated handwriting model in a database.

[1361] Step 6:

[1362] The user inputs a desired character string into an input field on the terminal.

[1363] Step 7:

[1364] The terminal transmits the input string and the user identification information to the server.

[1365] Step 8:

[1366] The server retrieves the corresponding handwriting model from the database based on the user identification information.

[1367] Step 9:

[1368] The server uses the acquired handwriting model to generate a handwriting imitation image that resembles the user's own handwriting for the input character string.

[1369] Step 10:

[1370] The server transmits the generated handwriting imitation image to the terminal.

[1371] Step 11:

[1372] The terminal displays the received handwriting imitation image on the user's screen.

[1373] Example 1

[1374] 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."

[1375] Conventional digital text generation methods have difficulty faithfully reproducing the handwriting of individual users, making it impossible to create unique digital content. Therefore, there is a demand for technology that can accurately reproduce a user's handwriting even in a digital environment and generate unique handwritten-style notes, letters, electronic signatures, etc.

[1376] 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.

[1377] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string that resembles the user's handwriting using the saved handwriting model, and means for transmitting the generated image to a user terminal and displaying it, thereby enabling the generation of digital text that accurately reproduces the user's handwriting.

[1378] "User input" refers to manual operations or data input by a user into the system.

[1379] A "handwritten character sample" is a digital image of a character handwritten by a user.

[1380] "Analysis" is the process of analyzing data and information in detail and extracting its meaning and characteristics.

[1381] "Handwriting features" are characteristic information about a particular handwriting, such as the thickness of the lines of handwritten characters, the pressure of the writing, the angle of the curves, and the order of strokes.

[1382] A "handwriting model" is a machine learning model that learns the user's unique handwriting characteristics.

[1383] "Learning" is the process by which machine learning algorithms understand patterns and features based on data.

[1384] "Storage" is the act of permanently recording data or information on a storage medium.

[1385] "Generating an image" means creating new image data using a specific algorithm.

[1386] A "user terminal" is a device (e.g., a smartphone, tablet, etc.) that a user uses to access the system.

[1387] "Display" refers to the act of visually presenting generated data or information to a user.

[1388] "Identification information" is data for uniquely identifying a user (for example, a user ID or authentication information).

[1389] "Image input means" refers to a device or function that captures a user's handwritten characters as a digital image.

[1390] "Image analysis" is the process of extracting information from digital images and analyzing it.

[1391] A "machine learning algorithm" is an algorithm that improves itself based on data.

[1392] A "database" is a system for efficiently managing, storing, and retrieving large amounts of data.

[1393] A "handwriting imitation image" is a digital image generated to resemble a user's own handwriting.

[1394] "Generation means" refers to the functions or devices that allow the system to create or convert new data or information.

[1395] A "prompt sentence" is text data that is input into an AI model.

[1396] The present invention provides a system that acquires a user's handwritten character sample, analyzes the handwriting characteristics of the sample to generate a machine learning model, and outputs an image of a character string input by the user that resembles the user's handwriting. Specific embodiments of the system are described below.

[1397] Acquiring handwriting data

[1398] A user writes characters by hand in an input field provided on a device such as a smartphone or tablet. When the user writes characters using a stylus pen or their finger, the device captures the characters as image data. The device then sends this image data and the user's identification information to a server. The hardware used is a typical smartphone or tablet, and the software has an image capture function built in.

[1399] Examples:

[1400] When a user inputs the characters "Good morning" by hand, the terminal captures the characters as an image and transmits it to the server together with the image data.

[1401] Training a handwriting model

[1402] The server analyzes the received image data and extracts handwriting features. This analysis uses an image processing library (e.g., OpenCV). Specifically, features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. Based on the extracted handwriting features, a machine learning algorithm (e.g., TensorFlow or PyTorch) is used to generate a handwriting model specific to the user, and this model is stored in a database.

[1403] Examples:

[1404] The server analyzes the image data of "Good morning" and extracts the characteristics of each character. For example, feature data such as the angle of the curve of "O" and the pressure of the stroke of "Wa" are generated, and this data is used to train a handwriting model of the user using a machine learning algorithm and saved.

[1405] Entering a string

[1406] The user enters a desired string on the device. For example, they enter "hello" into a text input field. The device then sends the string and user identification information to the server. The hardware used includes a typical smartphone or tablet input field.

[1407] Examples:

[1408] The user enters the string "hello" as text into the terminal, which then sends it to the server.

[1409] Generating handwriting imitation images

[1410] The server retrieves the corresponding handwriting model from a database based on the user's identification information. Using the retrieved model, the server generates an image of the input string that resembles the user's handwriting. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image.

[1411] Examples:

[1412] The server uses the text "hello" and the corresponding handwriting model to generate an image in which each character is converted to the user's own handwriting.

[1413] Viewing the generated results

[1414] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen, allowing the user to confirm that the input string appears similar to their own handwriting.

[1415] Examples:

[1416] The generated handwriting imitation image of "hello" is displayed on the user's device, allowing the user to confirm that the character "hello" is displayed in a form similar to their own handwriting.

[1417] Example prompt sentence:

[1418] "I would like to develop a system that acquires a user's handwriting, analyzes it using a machine learning model, and generates a handwritten-like image. Please tell me the specific steps: the user inputs characters into the device → the server analyzes the handwriting and generates a model → the input string is converted into a handwritten-like image and displayed. Please also tell me the technology and algorithms used."

[1419] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1420] Step 1:

[1421] The user writes handwritten characters on the device.

[1422] The user writes handwritten characters in an input field provided on the screen of a smartphone or tablet. The user inputs using a stylus pen or a finger. The device captures the handwritten characters as image data. The input data is an image of the handwritten characters, and the output data is the captured image file.

[1423] Specific behavior:

[1424] The user writes the words "Good morning" in the input field.

[1425] The device captures this text as an image and saves it in an internal temporary file.

[1426] Step 2:

[1427] The device captures the handwritten image and sends it to the server.

[1428] The device sends the captured image data and the user's identification information to the server. The input data is the image and user identification information acquired in the previous step, and the output data is the HTTP request sent to the server.

[1429] Specific behavior:

[1430] The device sends the captured image data of "Good morning" to the server as an HTTP POST request.

[1431] Step 3:

[1432] The server analyzes the image data and extracts handwriting features.

[1433] The server analyzes the received image data and uses an image processing library (e.g., OpenCV) to extract handwriting features. Features such as line thickness, pen pressure, curve angle, and stroke order are analyzed. The input data is an image file, and the output data is the extracted handwriting feature data.

[1434] Specific behavior:

[1435] The server reads the "Good morning" image file and converts it to grayscale.

[1436] The server quantifies the thickness of the lines and the angle of the curves and extracts this as feature data.

[1437] Step 4:

[1438] The server uses machine learning algorithms to generate and store handwriting models.

[1439] The server uses the extracted handwriting feature data to generate a user-specific handwriting model using a machine learning algorithm (e.g., TensorFlow or PyTorch). The generated model is stored in a database. The input data is the handwriting feature data, and the output data is the trained handwriting model.

[1440] Specific behavior:

[1441] The server uses a neural network to learn the feature data and generate a handwriting model for the user.

[1442] Save the generated model in the database.

[1443] Step 5:

[1444] The user enters a string into a text input field on the device.

[1445] The user inputs a desired string in text format into the terminal. The input data is the text string and user identification information, and the output data is an HTTP request sent to the server.

[1446] Specific behavior:

[1447] The user enters the string "Hello" into the text input field on the device.

[1448] The terminal transmits this text data and the user identification information to the server.

[1449] Step 6:

[1450] The server generates a handwriting imitation image using the user's handwriting model.

[1451] The server retrieves the corresponding handwriting model from the database based on the user's identification information. Using the retrieved model, it generates an image that resembles the user's handwriting from the input string. GAN or other image generation algorithms are used for image generation. The input data are the text string and the handwriting model, and the output data is a handwriting imitation image.

[1452] Specific behavior:

[1453] The server reads the corresponding user's handwriting model from the database.

[1454] Using the text string "hello" as input, a handwriting imitation image is generated using the GAN algorithm.

[1455] Step 7:

[1456] The server generates a handwriting imitation image and sends it to the terminal, which displays it.

[1457] The server sends the generated handwriting imitation image to the terminal, which displays the received image on the user's screen. The input data is the handwriting imitation image, and the output data is the displayed image.

[1458] Specific behavior:

[1459] The server sends the generated handwriting imitation image of "hello" to the terminal as an HTTP response.

[1460] The terminal receives this image, places it on the display area, and presents it to the user.

[1461] (Application example 1)

[1462] 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."

[1463] In conventional virtual stores, it was difficult for users to input messages in their own handwriting and have them reflected in customized products. This made it difficult to provide custom-made products that reflected each user's individuality in a digital environment, posing challenges in improving user satisfaction. Furthermore, existing systems lack a method for faithfully reproducing a user's handwriting, and there is a lack of technology for creating handwritten-style customized products.

[1464] 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.

[1465] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for learning and saving a handwriting model of the user based on the extracted handwriting features, means for receiving a character string input by the user and generating an image of the input character string by using the saved handwriting model to resemble the user's handwriting, means for adding a message that looks like the user's handwriting to customized products in the virtual store, and means for transmitting the generated image to a user terminal and displaying it. This enables users to easily create and order customized products with original designs using their own handwriting.

[1466] "User input" includes characters and information entered by a user using a device.

[1467] A "handwriting sample" is a character written by a user in their own handwriting on a device.

[1468] "Handwriting characteristics" refer to characteristics such as the thickness of lines, writing pressure, and angle of curves that are specific to the user's handwritten characters.

[1469] A "handwriting model" is a digital model generated using a machine learning algorithm based on the user's handwriting characteristics.

[1470] "Image generation" is a process of generating an image that resembles the user's handwriting using a handwriting model based on input text.

[1471] A "virtual store" is a virtual store operated on the Internet where users can select and purchase products.

[1472] A "customized product" is a product that is individually created by a user to add specific elements (in this case, a handwritten message) to suit their preferences.

[1473] "User identification information" is information for identifying a specific user, and typically includes a user ID and authentication information.

[1474] The "image input means" is a means for capturing characters written by a user on a device as an image.

[1475] The "generation means" is a means for converting a character string input by a user into a handwriting imitation image using a specific algorithm.

[1476] The "display means" is a means for displaying the generated handwriting imitation image on the user terminal.

[1477] MODE FOR CARRYING OUT THE INVENTION

[1478] System Configuration

[1479] The system for implementing this invention consists of a series of processes: a user inputs handwritten characters on a terminal, a handwriting model is generated for the user based on the input, and finally, the user's handwritten-style message is added to a customized product in a virtual store. The specific system configuration is described below.

[1480] Hardware Configuration

[1481] User terminals include smartphones, tablets, PCs, etc. These terminals are equipped with image input means (cameras, touch screens, etc.).

[1482] The server used is a cloud server, which has the capacity and database to run machine learning algorithms.

[1483] Software Configuration

[1484] The overall system utilizes the following software:

[1485] 1. Image analysis software: Using image processing libraries such as OpenCV, we analyze the user's handwritten character samples and extract handwriting features.

[1486] 2. Machine learning algorithms: Using TensorFlow and PyTorch, we generate a handwriting model of the user based on the extracted handwriting features.

[1487] 3. Generative Adversarial Network (GAN): Using a GAN, an image of the input string is generated that resembles the user's handwriting based on a model of the user's own handwriting.

[1488] 4. Image editing software: Use PIL (Python Imaging Library) to overlay the generated handwriting imitation image onto the customized product image.

[1489] Program processing

[1490] The server captures handwriting samples entered by the user on the device and transmits them to the server. The captured image is analyzed on the server side to extract handwriting features. Based on the extracted features, a machine learning algorithm is used to generate a handwriting model of the user and store it in a database.

[1491] When a user wishes to create a customized product, a request is sent from the terminal along with a character string. The server searches for a handwriting model based on the user's identification information and converts the input character string into a handwriting imitation image. The generated handwriting imitation image is overlaid on the customized product using the PIL library, and finally sent to the user's terminal for display.

[1492] Specific examples

[1493] For example, when a user writes "thank you" by hand using a smartphone, the character is captured as an image and sent to the server. The server analyzes the image and extracts handwriting features. Based on this, a handwriting model of the user is generated and stored in a database.

[1494] When a user customizes a T-shirt in the virtual store, they input the word "thank you" and send the text data to the server. The server uses a model of the user's handwriting to convert the input text into a handwriting imitation image and overlay this image on the T-shirt design. Finally, the generated T-shirt design is displayed on the user's smartphone.

[1495] Prompt Sentence Examples

[1496] When a user makes a request to the server for a customized product, the prompt text can be written as follows:

[1497] User ID: 12345, Text: "Thank you"

[1498] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1499] Step 1:

[1500] The user inputs handwritten characters on the device. The user writes characters such as "Thank you" by hand on the touch screen of a smartphone or tablet. This input is saved as image data on the device.

[1501] Input: Handwritten characters

[1502] Output: Image data

[1503] Step 2:

[1504] The device sends the captured image data to the server, along with the user's identification information, and uploads the image data to the server via the Internet.

[1505] Input: Image data, user identification information

[1506] Output: None (data transfer to server)

[1507] Step 3:

[1508] The server analyzes the received image data and extracts handwriting features, such as line thickness, pen pressure, and curve angle, using an image processing library such as OpenCV, to extract numerical data.

[1509] Input: Image data

[1510] Output: Handwriting features (numerical data)

[1511] Step 4:

[1512] The server uses a machine learning algorithm to generate a handwriting model of the user based on the extracted handwriting features and stores it in a database. TensorFlow and PyTorch are used to learn the handwriting features and generate a model specific to the user.

[1513] Input: Handwriting features

[1514] Output: Handwritten model (model data)

[1515] Step 5:

[1516] When a user creates a customized product in the virtual store, the user inputs an input string (e.g., "custom message") into the terminal and transmits this data to the server. The terminal then transmits the text data entered by the user and the user identification information to the server.

[1517] Input: Text data (input string), user identification information

[1518] Output: None (data transfer to server)

[1519] Step 6:

[1520] The server searches for and retrieves the corresponding handwriting model from the database based on the user identification information, searches for the saved handwriting model, and loads the corresponding model into memory.

[1521] Input: User Identification Information

[1522] Output: Handwritten model (model data)

[1523] Step 7:

[1524] The server uses a handwriting model to generate an image that resembles the user's handwriting. A generative adversarial network (GAN) is used to convert the input text into a handwriting-imitating image.

[1525] Input: Text data, handwriting model

[1526] Output: Handwriting imitation image

[1527] Step 8:

[1528] The server overlays the generated handwriting imitation image onto the customized product image. Using PIL, the handwriting imitation image is synthesized onto the customized product design image.

[1529] Input: Handwriting imitation image, customized product image

[1530] Output: Customized product image (after compositing)

[1531] Step 9:

[1532] The server transmits the synthesized customized product image to the user terminal, and the generated customized product image is transmitted to the user terminal via the Internet and displayed.

[1533] Input: Customized product image (after compositing)

[1534] Output: None (data transfer to user terminal)

[1535] Step 10:

[1536] The user terminal displays the received customized product image to the user, and the image is displayed on the terminal display so that the user can check the product customized with their own handwriting.

[1537] Input: Customized product image (after receiving)

[1538] Output: User confirms customized product image

[1539] 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.

[1540] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[1541] 1. Acquisition of handwriting data and emotion data

[1542] Program processing:

[1543] The user writes characters by hand in an input field on the device. The device captures the handwritten characters as image data. The device also simultaneously captures the user's voice and facial expressions while writing using a camera and microphone. The acquired voice and facial expression data are sent to an emotion engine, which analyzes the user's emotional state. The image data and emotional state information are then sent to the server.

[1544] Examples:

[1545] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[1546] 2. Learning handwriting and emotion models

[1547] Program processing:

[1548] The server analyzes the received image data and extracts handwriting features, such as pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate a handwriting model and emotion model specific to the user and stores them in a database.

[1549] Examples:

[1550] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[1551] 3. Entering text and emotions

[1552] Program processing:

[1553] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[1554] Examples:

[1555] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[1556] 4. Generating handwriting imitation images

[1557] Program processing:

[1558] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[1559] Examples:

[1560] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[1561] 5. Displaying the generated results

[1562] Program processing:

[1563] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[1564] Examples:

[1565] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[1566] This invention allows us to faithfully reproduce the handwriting and emotional state of individual users in a digital environment, thereby adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, etc.

[1567] The processing flow will be explained below.

[1568] Step 1:

[1569] The user writes characters by hand in an input field on the terminal.

[1570] The device captures the handwritten characters as image data.

[1571] Step 2:

[1572] The terminal transmits the captured handwritten character image and user identification information to the server.

[1573] Step 3:

[1574] The device captures the user's voice and facial expressions through a camera and microphone and sends them to the emotion engine, which analyzes the user's emotional state and sends the results to the server.

[1575] Step 4:

[1576] The server analyzes the received image data and extracts handwriting characteristics, such as line thickness, pressure, angle, and stroke order.

[1577] Step 5:

[1578] The server generates an emotion model based on the emotion state information received from the emotion engine, and the emotion model includes adjustment parameters for handwriting features according to the user's emotion state.

[1579] Step 6:

[1580] Based on the extracted handwriting features and emotion model, the server uses a machine learning algorithm to generate a handwriting model specific to the user and stores it in a database.

[1581] Step 7:

[1582] The user inputs a desired character string on the terminal.

[1583] The terminal transmits the character string, the user identification information, and the emotional state to the server.

[1584] Step 8:

[1585] The server retrieves a corresponding handwriting model from the database based on the user identification information, and further retrieves an emotion model based on the emotion state information.

[1586] Step 9:

[1587] The server uses the acquired handwriting model and emotion model to generate a handwriting imitation image of the input string that resembles the user's own handwriting. The style and angle of the characters are adjusted depending on the user's emotional state.

[1588] Step 10:

[1589] The server transmits the generated handwriting imitation image to the terminal.

[1590] Step 11:

[1591] The terminal displays the received handwriting imitation image on the user's screen.

[1592] For example, when a user writes "good morning" on a device, the device captures the characters as image data and sends it to the server. If the server recognizes that the user is in a "joy" emotional state, it performs analysis based on this information. When the user then types "hello," the server generates a handwriting imitation image with slightly thicker lines and softer curves, in line with the "joy" emotional state, and displays it on the device.

[1593] This invention makes it possible to generate character images that reflect the handwriting and emotions of individual users even in a digital environment, thereby imparting individuality and emotion to digital text. This opens up a wide range of applications, including handwritten-style notes, letters, and electronic signatures.

[1594] Example 2

[1595] 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."

[1596] Conventional systems can reproduce the characteristics of a user's handwriting when digitizing the user's handwritten text, but they are unable to reflect the user's emotional state. This makes it difficult to impart the user's personality and emotions to the digital text, resulting in a loss of the warmth and emotional expression inherent in handwritten text. The present invention aims to solve this problem by faithfully reproducing the user's handwriting and emotions in a digital environment.

[1597] 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.

[1598] In this invention, the server includes means for acquiring a handwritten character sample input by a user, means for analyzing the acquired handwritten character sample to extract handwriting features, means for analyzing the extracted handwriting features and emotional state to generate and save a handwriting model and emotional model dedicated to the user, means for receiving a character string input by the user and the emotional state and using the saved handwriting model and emotional model to generate an image that resembles the input character string and reflects the user's emotional state, and means for transmitting the generated image to a user terminal and displaying it. This makes it possible to generate a handwriting imitation image that reflects the user's individuality and emotions even in a digital environment.

[1599] The "means for acquiring handwritten character samples input by a user" refers to a device or method for capturing handwritten characters input by a user using a touch screen or a stylus pen and acquiring them as digital data.

[1600] The "means for extracting handwriting characteristics" refers to a device or method that analyzes and extracts characteristics such as writing pressure, line thickness, angle, and curve from an acquired handwritten character sample.

[1601] "Means for capturing emotional data" refers to a device or method that uses a camera or microphone to capture the user's facial expressions and voice and obtain them as digital data.

[1602] "Means for analyzing emotional state" refers to a device or method that analyzes captured facial expression and voice data to determine the emotional state of the user.

[1603] "Means for generating and storing handwriting models and emotion models" refers to a device or method that performs machine learning based on extracted handwriting features and analyzed emotional states, generates handwriting models and emotion models specific to the user, and stores them in a database.

[1604] The "means for generating an image that resembles the user's handwriting from an input character string and reflects the user's emotional state" refers to a device or method for generating a handwriting imitation image that resembles the user's handwriting, based on the character string and emotional state input by the user, using a stored handwriting model and emotional model.

[1605] The "means for transmitting the generated image to the user terminal and displaying it" refers to a device or method for transmitting the generated handwriting imitation image to the user terminal and displaying it on the screen of the terminal.

[1606] MODE FOR CARRYING OUT THE INVENTION

[1607] The present invention is a system that outputs an input character string as an image that resembles the user's handwriting, based on a sample of the user's handwriting and emotional state. Specific embodiments of the system will be described below.

[1608] Acquiring handwriting and emotion data

[1609] The user writes characters by hand in the input field on the device. The device captures this handwritten character as image data. Specifically, the device captures the user writing "good morning" on the touch screen with a stylus pen or finger and saves it as image data. The device also simultaneously captures the user's facial expressions and voice as they write using a camera and microphone. The acquired voice and facial expression data is sent to an emotion engine, which analyzes the user's emotional state. The device then sends this data to a server.

[1610] Examples:

[1611] The user writes "Good morning" by hand into the device, and the device captures the user's facial expression and voice (e.g., a cheerful voice or a smile). The server receives the image data of "Good morning" and the user's cheerful state (emotional state is "joy").

[1612] Learning handwriting and emotion models

[1613] The server analyzes the received image data and extracts handwriting features, specifically pen pressure, line thickness, angle, and curves. At the same time, the emotion engine analyzes the user's emotional state and provides this information to the server. Based on this information, the server uses machine learning algorithms to generate handwriting and emotion models specific to the user and stores them in a database. Machine learning libraries used include TensorFlow and PyTorch.

[1614] Examples:

[1615] The server analyzes the writing pressure and line thickness from the image data of "Good morning," and simultaneously receives the state of "happiness" from the emotion engine. Based on this, it learns the handwriting characteristics when the user is in a "happiness" state and creates a model.

[1616] String and emotion input

[1617] The user inputs a desired character string and their emotional state on the device. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The device then transmits the input character string and their emotional state to the server.

[1618] Examples:

[1619] The user inputs "hello" and selects (or recognizes) the emotional state. For example, the user can input "happy" in the text box, or the device can recognize the "happy" state from facial expressions and voice. This information is sent to the server.

[1620] Generating handwriting imitation images

[1621] The server retrieves the corresponding handwriting and emotion models from a database based on the user's identification information and emotional state. Using the retrieved models, it generates a handwriting imitation image that resembles the user's own handwriting and matches the user's emotional state. Generative adversarial networks (GANs) and other image generation algorithms are used to generate the image. Filtering and style changes are also applied according to the emotional state.

[1622] Examples:

[1623] The server uses the handwriting model and emotion model corresponding to the character string "hello" and the emotional state "happy," and generates a handwriting imitation image with thicker lines and slightly upward tilt.

[1624] Viewing the generated results

[1625] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen.

[1626] Examples:

[1627] The generated handwriting imitation image of "hello" is displayed on the user's device. The image looks like the user wrote it in a "happy" state.

[1628] Example prompt sentence:

[1629] A user writes "Good morning" in an input field by hand, and their voice (cheerful voice) and facial expression (happy smile) are captured to analyze their emotional state. Next, the same user types "Hello" and selects "Happy" as their emotional state. Using this information, the server should generate a handwriting imitation image of "Hello."

[1630] By using these steps, we can faithfully reproduce the user's handwriting and emotional state in a digital environment, adding personality and emotion to digital text. This system can be applied to handwritten notes, letters, electronic signatures, and more.

[1631] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1632] Step 1: User inputs a sample of their own handwriting

[1633] The user writes characters by hand using a stylus pen or their finger in the input field of the device. The device acquires this input and saves it as digital image data. A specific input is a string of characters such as "Good morning." The output is image data of the string of characters.

[1634] Step 2: Capture handwriting and emotion data

[1635] The device uses a camera and microphone to capture the user's facial expressions and voice while they are writing. The camera and microphone signals are processed to obtain facial expression data and voice data. Specifically, the device captures the user's smile and cheerful voice while they are writing "Good morning." The output is facial expression data and voice data.

[1636] Step 3: Analyze emotional state

[1637] The device sends the captured facial expression and voice data to the emotion engine, which analyzes the data and determines the user's emotional state. The specific inputs are facial expression and voice data, and the output is emotional state information such as "joy."

[1638] Step 4: Sending data

[1639] The terminal transmits image data of the handwritten characters and the analyzed emotional state to the server. The specific inputs are the image data and the emotional state information, and the output is the data arriving at the server.

[1640] Step 5: Extracting handwriting features

[1641] The server analyzes the received image data and extracts handwriting features such as pressure, line thickness, angle, and curve. Specifically, it uses an image processing algorithm to extract these features from the input data (handwriting image). The output is the extracted handwriting feature data.

[1642] Step 6: Generate emotion and handwriting models

[1643] The server applies machine learning algorithms to the extracted handwriting features and emotional state information to generate a user-specific handwriting model and emotion model. Specifically, it uses machine learning libraries such as TensorFlow and PyTorch. The input is the handwriting feature data and emotional state information, and the output is the generated model.

[1644] Step 7: Save the model

[1645] The server stores the generated handwriting and emotion models in a database. Specifically, it serializes the models and writes them to the database. The input is the model data, and the output is the results stored in the database.

[1646] Step 8: Entering Strings and Emotional States

[1647] The user inputs a desired string (e.g., "hello") and emotional state on the terminal. The emotional state is acquired using text input, voice input, facial expression recognition, etc. The input is the string and emotional state information. The output is the input data on the terminal.

[1648] Step 9: Sending data (input text and emotional state)

[1649] The terminal sends the input string and emotional state to the server. The input is the string and emotional state information, and the output is the data arriving at the server.

[1650] Step 10: Getting the Model

[1651] The server retrieves the corresponding handwriting model and emotion model from the database. The input is the user identification information, and the output is the model data.

[1652] Step 11: Generate handwriting imitation images

[1653] The server uses the acquired model to generate a handwriting imitation image that matches the user's handwriting and emotional state. Specifically, it uses a generative adversarial network (GAN). The input is the string of characters, emotional state information, and model data, and the output is a handwriting imitation image.

[1654] Step 12: Sending the generated results

[1655] The server sends the generated handwriting imitation image to the terminal. The input is the handwriting imitation image, and the output is the image data arriving at the terminal.

[1656] Step 13: Display on the screen

[1657] The terminal displays the received handwriting imitation image on the user's screen. The input is the handwriting imitation image data, and the output is the display result on the user's screen.

[1658] (Application example 2)

[1659] 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."

[1660] Conventional systems generate handwriting imitation images based only on a user's handwriting samples, making it impossible to reflect changes in emotions or mood. Furthermore, there was a lack of a way to visually convey individual customers' emotions and personalities when improving customer experiences and promoting them in physical stores. In particular, there were limited ways for customers to share their emotions and feedback with other customers in the store. This made it difficult to provide an engaging customer experience using digital signage and displays.

[1661] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1662] In this invention, the server includes means for acquiring handwritten character samples and emotional states input by a user, means for analyzing the acquired handwritten character samples and emotional states to extract handwriting features and emotional features, and means for learning and saving a handwriting model and emotional model of the user based on the extracted handwriting features and emotional features, thereby enabling the generation of handwriting imitation images that reflect the user's individuality and emotions.

[1663] A "handwritten character sample" is a sample of characters input by handwriting by the user.

[1664] "Emotional state" is information that indicates the user's emotional or psychological state.

[1665] "Handwriting characteristics" refers to characteristics such as the shape, writing pressure, line thickness, angle, and curve of the user's handwritten characters.

[1666] "Emotional features" refer to features such as facial expressions, voice, and writing style based on the user's emotional state.

[1667] A "handwriting model" is a model that reproduces a user's unique handwriting, generated by a machine learning algorithm based on a user's handwriting samples.

[1668] The "emotion model" is a model for analyzing the user's emotional state and generating a handwriting imitation image that reflects the emotion.

[1669] "Image input means" refers to means for obtaining handwritten character samples using a camera, scanner, or the like.

[1670] The "emotion analysis means" is a means for analyzing voice and facial expression data to identify the emotional state of the user.

[1671] The "generating means" is a means for converting a character string input by a user into a handwriting imitation image using the stored handwriting model and emotion model.

[1672] The "database" is a storage device that stores the generated handwriting model and emotion model and makes them searchable as needed.

[1673] "User identification information" is information for uniquely identifying a user.

[1674] A "handwriting imitation image" is an image that reflects the user's handwriting and emotions, generated using a handwriting model and emotion model of the user.

[1675] Specific embodiments for carrying out the present invention will be described below.

[1676] The present invention is a system that generates handwritten-style message images based on a customer's handwriting sample and emotional state and displays them on an in-store display to improve customer experience in brick-and-mortar stores. This system generates a handwriting model and an emotion model based on user input data and uses them to generate a handwriting imitation image.

[1677] Hardware Configuration

[1678] Input device: Use an iPad or Android tablet, which allows customers to input handwritten messages.

[1679] Camera: Uses the device's built-in camera, which allows capturing customer facial expressions and analyzing their emotional state.

[1680] Microphone: Uses the device's built-in microphone, which allows us to capture the customer's voice and analyze their emotional state.

[1681] Server: Use a cloud server (e.g., AWS, Google Cloud) to process large amounts of data and generate models.

[1682] Display: A large display will be used in the store to display the generated handwritten message image.

[1683] Software Configuration

[1684] Handwriting capture: The device app combines a journal and note-taking app with the ability to accept handwriting input from customers.

[1685] Emotion analysis engine: Uses software (e.g., FaceAPI, Azure Cognitive Services) to capture voice and facial expression data and analyze emotions.

[1686] Data transmission: It has network functionality to transmit captured data to a cloud server in real time.

[1687] Model generation: On a cloud server, a handwriting model and an emotion model of the user are generated using a generative adversarial network (GAN) or a machine learning algorithm (e.g., TensorFlow, PyTorch).

[1688] Image generation & display: The generated handwriting imitation image is sent to the display via a web service and displayed in real time.

[1689] Specific examples

[1690] 1. Customer operations:

[1691] He writes in his own handwriting on a tablet inside the store, "I love this store!"

[1692] Cameras and microphones capture customers' facial expressions and voices.

[1693] 2. Emotion analysis:

[1694] The emotion analysis engine analyzes facial expressions and voice data to detect the emotional state of "joy."

[1695] 3. Data transmission:

[1696] The handwritten character sample and emotion data are sent to a cloud server.

[1697] 4. Model generation and image output:

[1698] The server analyzes handwriting features and emotion features based on the received data and generates a handwriting model and emotion model.

[1699] The sentence "I love this store!" is converted into a handwritten style that matches the emotion of joy, and is generated in slightly thicker strokes.

[1700] 5. Results display:

[1701] The generated handwriting imitation image is displayed on a large display inside the store.

[1702] The display features a hand-drawn image that reflects the customer's joyful emotions and reads, "We love this store!", for all customers to see.

[1703] Example prompts for generative AI models

[1704] "Generate a handwritten message image based on the handwritten characters and emotion data below.

[1705] Text: I love this store!

[1706] Emotion: Joy

[1707] Format: Handwritten, light background, thick strokes

[1708] This system will improve the customer experience in physical stores and make them feel more connected to other customers.

[1709] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1710] Step 1:

[1711] The device captures the user's handwritten text in the input field as image data, while the device's built-in camera and microphone simultaneously capture the user's facial expressions and voice, providing data for emotion analysis.

[1712] Input: Handwritten text, facial expression data, voice data

[1713] Output: Image data (handwritten characters), emotion analysis data (facial expressions, voice)

[1714] Step 2:

[1715] The device transmits the captured image data and sentiment analysis data to a server, which applies an image analysis algorithm to extract handwriting features from the image data and uses a sentiment analysis engine to extract sentiment features from the sentiment analysis data.

[1716] Input: Image data, emotion analysis data

[1717] Output: handwriting features, emotion features

[1718] Step 3:

[1719] The server uses the extracted handwriting features and emotion features to generate a user-specific handwriting model and emotion model using a machine learning algorithm (e.g., TensorFlow, PyTorch), and stores this in a database.

[1720] Input: handwriting features, emotion features

[1721] Output: Handwriting model, emotion model

[1722] Step 4:

[1723] The user inputs a desired character string and their emotional state on the terminal. The terminal transmits the data to the server. The server uses the stored handwriting model and emotion model to generate a handwriting imitation image that resembles the user's handwriting and corresponds to their emotional state.

[1724] Input: Input string, emotional state

[1725] Output: Handwriting imitation image

[1726] Step 5:

[1727] The server sends the generated handwriting imitation image to the terminal, which displays the image on the user's screen and also sends it to a large display in the store to share with other customers.

[1728] Input: Handwriting imitation image

[1729] Output: Images displayed on user devices and large displays

[1730] 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.

[1731] 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.

[1732] 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.

[1733] 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.

[1734] 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.

[1735] 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.

[1736] 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).

[1737] 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.

[1738] 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."

[1739] 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.

[1740] 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).

[1741] 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.

[1742] 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.

[1743] 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.

[1744] 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.

[1745] 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.

[1746] 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.

[1747] 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.

[1748] 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.

[1749] 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.

[1750] 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.

[1751] The following is further disclosed regarding the above embodiment.

[1752] (Claim 1)

[1753] means for acquiring a handwritten character sample input by a user;

[1754] A means for analyzing the acquired handwritten character sample and extracting handwriting features;

[1755] means for learning and storing a model of the user's handwriting based on the extracted handwriting features;

[1756] means for receiving a character string input by a user and generating an image of the input character string by using the stored handwriting model to resemble the user's handwriting;

[1757] means for transmitting the generated image to a user terminal and displaying it;

[1758] A system including:

[1759] (Claim 2)

[1760] an image input means for capturing a handwritten character sample input by a user;

[1761] A means for extracting handwriting features from the captured handwritten character sample by image analysis;

[1762] A means for generating a user-specific handwriting model using a machine learning algorithm based on the extracted handwriting features;

[1763] a means for storing the generated handwriting model in a database;

[1764] a generating means for receiving a character string input by a user and converting the input character string into a handwriting imitation image by applying the stored handwriting model;

[1765] means for transmitting the generated handwriting imitation image to a user terminal and displaying it;

[1766] 10. The system of claim 1, comprising:

[1767] (Claim 3)

[1768] a means for searching for a corresponding handwriting model using user identification information and converting the character string into a handwriting imitation image;

[1769] means for generating a corresponding handwriting imitation image based on an input character string and user identification information received from a user terminal, and transmitting the image to the user terminal;

[1770] 10. The system of claim 1, comprising:

[1771] "Example 1"

[1772] (Claim 1)

[1773] means for acquiring a handwritten character sample input by a user;

[1774] A means for analyzing the acquired handwritten character sample and extracting handwriting features;

[1775] means for learning and storing a model of the user's handwriting based on the extracted handwriting features;

[1776] means for receiving a character string input by a user and generating an image of the input character string by using the stored handwriting model to resemble the user's handwriting;

[1777] means for transmitting the generated image to a user terminal and displaying it;

[1778] A system including:

[1779] (Claim 2)

[1780] an image input means for capturing a handwritten character sample input by a user;

[1781] A means for extracting handwriting features from the captured handwritten character sample by image analysis;

[1782] A means for generating a user-specific handwriting model using a machine learning algorithm based on the extracted handwriting features;

[1783] a means for storing the generated handwriting model in a database;

[1784] a generating means for receiving a character string input by a user and converting the input character string into a handwriting imitation image by applying the stored handwriting model;

[1785] means for transmitting the generated handwriting imitation image to a user terminal and displaying it;

[1786] 10. The system of claim 1, comprising:

[1787] (Claim 3)

[1788] a means for searching for a corresponding handwriting model using user identification information and converting the character string into a handwriting imitation image;

[1789] means for generating a corresponding handwriting imitation image based on an input character string and user identification information received from a user terminal, and transmitting the image to the user terminal;

[1790] 10. The system of claim 1, comprising:

[1791] "Application Example 1"

[1792] (Claim 1)

[1793] means for acquiring a handwritten character sample input by a user;

[1794] A means for analyzing the acquired handwritten character sample and extracting handwriting features;

[1795] means for learning and storing a model of the user's handwriting based on the extracted handwriting features;

[1796] means for receiving a character string input by a user and generating an image of the input character string by using the stored handwriting model to resemble the user's handwriting;

[1797] a means for adding a user's handwritten message to a customized product in the virtual store;

[1798] means for transmitting the generated image to a user terminal and displaying it;

[1799] A system including:

[1800] (Claim 2)

[1801] an image input means for capturing a handwritten character sample input by a user;

[1802] A means for extracting handwriting features from the captured handwritten character sample by image analysis;

[1803] A means for generating a user-specific handwriting model using a machine learning algorithm based on the extracted handwriting features;

[1804] a means for storing the generated handwriting model in a database;

[1805] a generating means for receiving a character string input by a user and converting the input character string into a handwriting imitation image by applying the stored handwriting model;

[1806] a means for superimposing the generated handwriting imitation image on customized products in a virtual store;

[1807] means for transmitting the generated handwriting imitation image to a user terminal and displaying it;

[1808] 10. The system of claim 1, comprising:

[1809] (Claim 3)

[1810] a means for searching for a corresponding handwriting model using user identification information and converting the character string into a handwriting imitation image;

[1811] a means for generating a corresponding handwriting imitation image based on the input character string and user identification information received from the user terminal, superimposing the image on a customized product in the virtual store, and transmitting the image to the user terminal;

[1812] 10. The system of claim 1, comprising:

[1813] "Example 2: Combining Emotion Engines"

[1814] (Claim 1)

[1815] means for acquiring a handwritten character sample input by a user;

[1816] A means for analyzing the acquired handwritten character sample and extracting handwriting features;

[1817] means for analyzing the extracted handwriting features and emotional state to generate and store a user-specific handwriting model and emotional model;

[1818] means for receiving a character string and an emotional state input by a user, and generating an image that resembles the user's handwriting and reflects the emotional state by using the stored handwriting model and emotional model;

[1819] means for transmitting the generated image to a user terminal and displaying it;

[1820] A system including:

[1821] (Claim 2)

[1822] image input means for capturing handwritten character samples input by a user and means for capturing emotion data;

[1823] A means for extracting handwriting features from the captured handwritten character sample by image analysis and analyzing emotion data to determine the emotional state;

[1824] means for generating a user-specific handwriting model and emotion model using a machine learning algorithm using the extracted handwriting features and emotion state;

[1825] a means for storing the generated handwriting model and emotion model in a database;

[1826] a generating means for receiving a character string and an emotional state input by a user, and converting the input character string into a handwriting imitation image by applying the stored handwriting model and emotional model;

[1827] means for transmitting the generated handwriting imitation image to a user terminal and displaying it;

[1828] 10. The system of claim 1, comprising:

[1829] (Claim 3)

[1830] a means for searching for a corresponding handwriting model and emotion model using user identification information, and converting the character string and emotion state into a handwriting imitation image;

[1831] means for generating a corresponding handwriting imitation image based on an input character string, user identification information, and emotional state received from a user terminal, and transmitting the image to the user terminal;

[1832] 10. The system of claim 1, comprising:

[1833] "Application example 2 when combining emotion engines"

[1834] (Claim 1)

[1835] means for acquiring user-input handwriting samples and emotional states;

[1836] A means for analyzing the acquired handwritten character sample and emotional state to extract handwriting features and emotional features;

[1837] means for learning and storing a user's handwriting model and emotion model based on the extracted handwriting features and emotion features;

[1838] a means for receiving a character string and an emotional state input by a user, and generating a handwriting imitation image according to the emotional state by making the input character string resemble the user's handwriting using the stored handwriting model and emotional model;

[1839] means for transmitting the generated image to a user terminal and displaying it;

[1840] A system including:

[1841] (Claim 2)

[1842] image input and emotion analysis means for capturing user-input handwriting samples and emotional states;

[1843] A means for extracting handwriting features and emotional features by analyzing the captured handwritten character sample and the emotional state;

[1844] A means for generating a user-specific handwriting model and emotion model using a machine learning algorithm based on the extracted handwriting features and emotion features;

[1845] a means for storing the generated handwriting model and emotion model in a database;

[1846] a generating means for receiving a character string and an emotional state input by a user, and converting the input character string into a handwriting imitation image by applying the stored handwriting model and emotional model;

[1847] means for transmitting the generated handwriting imitation image to a user terminal and displaying it;

[1848] 10. The system of claim 1, comprising:

[1849] (Claim 3)

[1850] a means for searching for a corresponding handwriting model and emotion model using the user identification information and the emotion state, and converting the character string into a handwriting imitation image corresponding to the emotion state;

[1851] means for generating a corresponding handwriting imitation image based on an input character string, an emotional state, and user identification information received from a user terminal, and transmitting the image to the user terminal;

[1852] 10. The system of claim 1, comprising: [Explanation of symbols]

[1853] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for acquiring a handwritten character sample input by a user; A means for analyzing the acquired handwritten character sample and extracting handwriting features; means for learning and storing a model of the user's handwriting based on the extracted handwriting features; means for receiving a character string input by a user and generating an image of the input character string by using the stored handwriting model to resemble the user's handwriting; means for transmitting the generated image to a user terminal and displaying it; A system including:

2. an image input means for capturing a handwritten character sample input by a user; A means for extracting handwriting features from the captured handwritten character sample by image analysis; A means for generating a user-specific handwriting model using a machine learning algorithm based on the extracted handwriting features; a means for storing the generated handwriting model in a database; a generating means for receiving a character string input by a user and converting the input character string into a handwriting imitation image by applying the stored handwriting model; means for transmitting the generated handwriting imitation image to a user terminal and displaying it; The system of claim 1 , comprising:

3. a means for searching for a corresponding handwriting model using user identification information and converting the character string into a handwriting imitation image; means for generating a corresponding handwriting imitation image based on an input character string and user identification information received from a user terminal, and transmitting the image to the user terminal; The system of claim 1 , comprising:

Images