system

Abstract

We provide the system. [Solution] Means for obtaining images of returned items, A means for analyzing the acquired images and extracting the characteristics of damage to the article, A means for comparing the extracted features with previously registered damage data, A means of determining whether compensation is applicable based on the results of the above comparison, Means for notifying the aforementioned judgment result, The notification means includes means for providing results in real time via a mobile terminal device, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] The damage assessment for the returned device has been conventionally manual, which has problems of taking time, being prone to subjectivity, and lacking consistency. Such an assessment process may reduce business efficiency and cause inappropriate compensation assessment. To solve these problems, an automated system for fairly and quickly assessing damage is required.

Means for Solving the Problems

[0005] To address the above challenges, the present invention acquires images of returned equipment and uses AI to analyze the images and extract damage features. The extracted features are automatically compared with past data to provide a system that determines whether compensation is necessary. This system includes image acquisition means, feature extraction means, data comparison means, and notification means, thereby enabling a fair and consistent evaluation to be performed quickly.

[0006] "Image acquisition means" refers to means for photographing the external appearance of the returned device and generating image data from that photograph.

[0007] "Image analysis means" refers to means for extracting features related to damage to the device from acquired images.

[0008] "Feature extraction" is the process of identifying specific patterns and shapes of damaged areas contained in an image and extracting their characteristic features.

[0009] A "data comparison method" is a means of comparing the extracted damage characteristics with existing data in a database that has been registered in the past.

[0010] "Decision-making means" refers to the means used to determine whether or not damage is covered by compensation, based on image analysis and data comparison results.

[0011] A "notification method" is a means of communicating the decision result to the user.

[0012] A "deep learning model" is an AI technology used to extract high-dimensional features from complex image data.

[0013] "Damage assessment" is the process of analyzing the extent and location of damage to the equipment and calculating the compensation amount based on the results. [Brief explanation of the drawing]

[0014] [Figure 1]It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0020] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the 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.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0028] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] The image analysis system of the present invention is built to automate the process of efficiently evaluating returned equipment and analyzing its damage. Specific embodiments of this system are described below.

[0036] Upon arrival at the office, the returned device is placed in a dedicated shooting booth. This booth is equipped with a high-performance camera that captures images of the device from multiple angles. These images are immediately transferred to a server, and the subsequent analysis process begins.

[0037] The server receives the transmitted images and performs analysis within its AI image analysis module. This module utilizes deep learning technology to automatically extract damage features such as cracks and surface scratches from the images. This sophisticated process allows for accurate detection of even minute damage.

[0038] Next, the server compares the feature extraction results with the database. The database contains a collection of past damage data and corresponding reports, and uses this to calculate the similarity of the damage. If the similarity exceeds a certain threshold, the server determines that the damage is eligible for compensation.

[0039] Finally, the server notifies the user of the decision. This notification is sent via email or a dedicated application, and a detailed analysis report is attached to the result. Based on this report, the user can then initiate the compensation process.

[0040] As a concrete example, consider a case where a device is returned and a large crack is found on its screen. The server compares the shape of the crack with its similarity to past data and determines that it is human-caused damage. It then automatically calculates the compensation amount and notifies the user. This entire process ensures operational efficiency and fair judgment.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The terminal is placed inside the shooting booth, and a high-resolution camera captures images from multiple angles of the exterior. The captured images are then transmitted directly to the server in digital format.

[0044] Step 2:

[0045] The server inputs the received image into the AI ​​image analysis module and starts the image analysis process. In this process, the server uses deep learning technology to extract damage features such as scratches and cracks.

[0046] Step 3:

[0047] The server compares the extracted features with historical damage data stored in its internal database. Here, it calculates the similarity between the current features and similar damage patterns reported in the past.

[0048] Step 4:

[0049] The server determines whether the damage is covered by compensation based on the similarity calculation results. This determination applies pre-defined criteria to determine whether compensation is necessary depending on the type and extent of the damage.

[0050] Step 5:

[0051] The server notifies the user of the decision. The notification is sent via email or a dedicated application, and a detailed analysis report is attached. This allows the user to immediately initiate the necessary compensation procedures.

[0052] (Example 1)

[0053] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0054] In assessing damage to returned equipment, conventional methods often rely heavily on human judgment, limiting the consistency and speed of evaluations. Furthermore, there is a need for highly accurate analysis of diverse damage patterns. Additionally, there is a lack of effective means to communicate damage assessment results to users, highlighting the need for increased efficiency and fairness in decision-making.

[0055] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0056] In this invention, the server includes means for installing returned equipment and acquiring images from multiple angles in a controlled environment; means for automatically extracting damage features of the equipment using a deep learning algorithm; and means for comparing the acquired damage features with past data and evaluating the similarity. This enables accurate and rapid damage assessment and fair compensation decisions while minimizing human intervention.

[0057] "Returned equipment" refers to devices that have been used by a user and then returned to their original location.

[0058] A "controlled shooting environment" refers to a shooting booth configuration where the light source and shooting conditions are standardized, minimizing the influence of the external environment.

[0059] A "deep learning algorithm" refers to a technology that automatically learns specific features from images through training with large amounts of data, and performs damage detection and classification.

[0060] "Damage characteristics" refer to information about the location, shape, and size of cracks and damage extracted through image analysis.

[0061] "Similarity assessment" refers to the process of comparing extracted damage features with data in past databases and quantifying their correlation.

[0062] A "communication network" refers to a network infrastructure built for the purpose of exchanging data.

[0063] "User" refers to the final recipient of a system or service.

[0064] This invention provides a system for automatically evaluating the damage to returned equipment and efficiently determining whether compensation is applicable. A specific embodiment of this system is described below.

[0065] First, when a user returns a device to the office, it is placed in a dedicated shooting booth. The shooting booth provides a controlled environment, maintaining a constant position for the light source and subject, making it possible to acquire high-definition images every time, eliminating the influence of ambient light. This shooting booth is equipped with a high-performance digital camera that captures detailed images from multiple angles of the device.

[0066] Next, the server quickly receives the captured images and stores them in a central storage device. This data is immediately sent to an AI analysis module, where it is automatically processed using a deep learning algorithm. This algorithm, built using TENSORFLOW®, automatically extracts damage features and recognizes crack and damage patterns with high accuracy. This allows users to quickly detect even minute damage that might be overlooked by human visual inspection.

[0067] Subsequently, the server compares the extracted damage characteristics with past damage information stored in a database. This database records past damage cases, their causes, and repair history, and the server evaluates the similarity to this information to determine compensation. This evaluation result is standardized through an automated process, enabling fair and consistent decisions.

[0068] Ultimately, the server determines whether compensation is applicable and notifies the user of the result. This notification is sent via email or a dedicated application, allowing the user to check the result immediately. The notification includes an analysis report detailing the damage, which the user can use to proceed with the subsequent procedures.

[0069] For example, if a user returns a device with a cracked screen, the server analyzes the shape of the crack by comparing it to past data and determines that it is human-caused damage. At this point, the server automatically calculates the compensation amount and notifies the user. This process allows users to receive compensation quickly and fairly.

[0070] An example of a prompt when using a generative AI model is: "Please analyze the damage to the returned device in detail and inform us about the possibility of compensation."

[0071] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0072] Step 1:

[0073] Upon returning the device, the user places it in a dedicated shooting booth. This booth eliminates ambient light and uses a high-performance camera to capture images of the device from multiple angles. The input data consists of the captured high-resolution images of the device. This results in clear image data being obtained as output.

[0074] Step 2:

[0075] The server receives captured images via the network and stores them in a central repository. The input consists of multiple image files received by the server. The server organizes these image data appropriately and prepares them for immediate use in the subsequent analysis process. The output of this step is the image data ready for analysis.

[0076] Step 3:

[0077] The server passes the saved image to the AI ​​image analysis module and starts the analysis. The input is image data ready for analysis. The server uses a deep learning algorithm powered by TensorFlow to automatically extract damage features such as cracks and scratches from the image. The output includes the extracted damage features as data.

[0078] Step 4:

[0079] The server compares the extracted damage features with a database of past damages and calculates similarity. The inputs are the extracted damage features and the database of past damages. The server uses statistical methods to evaluate the similarity and establish criteria for determining whether compensation is necessary. The output of this step is the damage similarity score.

[0080] Step 5:

[0081] The server makes a compensation decision based on the similarity score and notifies the user of the result. The input is the damage similarity score. The server generates the result and reports it to the user via email or a dedicated application. The output is a notification of the decision result, including a detailed analysis report.

[0082] (Application Example 1)

[0083] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0084] In the damage assessment process for returned goods at logistics centers, there is a need to reduce the variability of human judgment and automate the process for quick and fair compensation decisions. Furthermore, building a system that improves work efficiency by providing real-time notification of results is a key challenge.

[0085] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0086] In this invention, the server includes means for acquiring images of returned items, means for analyzing the acquired images and extracting characteristics of the damage to the items, means for comparing the extracted characteristics with previously registered damage data, means for determining whether the items are eligible for compensation based on the results of the comparison, means for notifying the determination result, and means for the notification means to provide the result in real time via a mobile terminal device. This makes it possible to automate the damage assessment of returned goods at a logistics center and to quickly make fair compensation decisions.

[0087] "Returned items" refers to all products and goods that have been sent back after being used by a customer or user.

[0088] "Means of acquiring images" refers to devices and technologies for visually capturing the whole or parts of an object and saving them as digital data.

[0089] "Means for analyzing images and extracting the characteristics of damage to an item" refers to technologies and devices that process acquired image data to identify damage to the surface or interior of an item and automatically recognize its detailed characteristics.

[0090] "Means of comparing with previously registered damage data" refers to technologies and processes for matching information on previously recorded damage with newly identified damage characteristics.

[0091] "Means of determining whether an item is eligible for compensation" refers to criteria or devices used to determine whether an item meets the compensation requirements, based on the characteristics of the damage and past data.

[0092] "Means of notifying the decision result" refers to communication technologies and systems used to convey information to relevant parties regarding whether compensation is possible and details of the damage.

[0093] "Means of providing results in real time via mobile terminal devices" refers to technologies and systems that allow users to receive decision results immediately through portable digital devices.

[0094] The system implementing this invention incorporates advanced image analysis technology to efficiently assess the damage to returned items.

[0095] First, the items returned by the user are placed in a dedicated shooting environment, and multiple images are acquired from the entire object and from the necessary angles using the camera of a mobile device. This mobile device has a dedicated application installed for transferring the images to a server in the cloud.

[0096] The server performs AI image analysis using the received image data. Specifically, a model utilizing deep learning technology automatically extracts damage features and compares this information with past damage data stored in a database. The AI ​​model employs machine learning techniques using TensorFlow.

[0097] The server determines the need for compensation based on the matching results. If compensation is deemed necessary, the server notifies the user via the mobile device. The notification includes details of the damage and information regarding the application of compensation. The results are provided in real time using the push notification technology used.

[0098] As a concrete example, consider a case where a household electrical appliance returned to a logistics center has a cracked screen due to a fall. This system identifies the damage in real time and immediately determines whether it is eligible for compensation by comparing it with past data.

[0099] An example of a prompt message might be: "From this image, detect the extent of damage to the product upon return and report it by comparing it to similar data." This inquiry is then posed to the generating AI model.

[0100] This will not only dramatically improve operational efficiency at logistics centers, but also ensure fairness and speed in compensation.

[0101] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0102] Step 1:

[0103] The user photographs the returned item using the camera on their mobile device. The user photographs the entire item and any suspected damaged areas from multiple angles and inputs the image data through a dedicated application on the device. The input image data is then transferred to a server in the cloud.

[0104] Step 2:

[0105] The server passes the received image data to the AI ​​analysis module. The server uses a generative AI model based on TensorFlow to analyze the damage features of the object from the image. Here, information extracted as damage features is output in response to the image data input. Specifically, it performs data calculations to detect crack and deformation patterns.

[0106] Step 3:

[0107] The server compares the extracted damage features with historical damage data stored in the database. The server receives the damage information from the analysis as input and calculates its similarity to past compensation cases. This results in an output determining whether the damage is eligible for compensation. Specifically, it performs data matching using a similarity calculation algorithm.

[0108] Step 4:

[0109] The server notifies the mobile device of the assessment result and communicates it to the user. The notification from the server includes detailed information about the damage and the assessment result regarding compensation, and is pushed to the user in real time. The output of the notification serves as a basis for the user to decide whether to file a compensation claim. Specifically, the result is sent via a notification API.

[0110] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0111] This invention provides a system that combines user emotion recognition with a device damage assessment process. This system not only analyzes the damage to returned devices using image analysis and makes compensation decisions based on the analysis results, but also recognizes the user's emotional state and optimizes the notification process.

[0112] First, the returned device is placed in a shooting booth, and images are automatically taken from multiple angles. The captured images are immediately sent to a server. This image data is analyzed by an AI image analysis module, and damage characteristics are extracted. The server compares the extracted characteristics with an internal historical database to determine whether the damage is eligible for compensation.

[0113] Subsequently, the server considers the user's emotions when notifying them of the decision. This function is provided by an emotion recognition module, which analyzes the user's voice and written feedback data. If the user's emotions are negative, such as dissatisfaction or surprise, the server adjusts the notification content to include a polite and convincing explanation before sending it.

[0114] For example, if a user is dissatisfied with device damage, emotion recognition can detect that emotion. In this case, the server sends a report in addition to the usual notification, which includes a detailed explanation of the compensation and information on how the damage was assessed. This demonstrates consideration for the user and improves the customer experience.

[0115] This invention makes it possible to provide higher quality service not only by accurately assessing damage, but also by incorporating user emotional responses.

[0116] The following describes the processing flow.

[0117] Step 1:

[0118] Once the device is returned, it is placed in the shooting booth. A high-resolution camera captures images of various parts of the device from multiple angles, and the acquired images are sent to a server as digital data.

[0119] Step 2:

[0120] The server inputs the received image data into an AI image analysis module. This module extracts damage features such as scratches and cracks from the image. The extracted features are detailed data indicating the type, location, and extent of the damage.

[0121] Step 3:

[0122] The server compares the extracted damage characteristics with a historical database and calculates the similarity. The database contains previously recorded damage cases, and the server uses this comparison to determine whether the damage is eligible for compensation.

[0123] Step 4:

[0124] Based on the decision, the server creates notification content while considering the user's emotional state. The emotion recognition module analyzes the voice data and feedback obtained from the user to determine the user's emotions.

[0125] Step 5:

[0126] The server responds flexibly to the user's emotional state. For example, if negative emotions are detected, the notification will include additional information about compensation details and the decision-making process, demonstrating consideration for the user.

[0127] Step 6:

[0128] Finally, the server sends the assessment results and adjusted notification to the user. This allows the user to accurately understand the damage assessment results and compensation details and proceed with the necessary procedures.

[0129] (Example 2)

[0130] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0131] Conventional damage assessment systems rely solely on objective data regarding the condition of returned equipment when determining damage, which means they cannot take into account user emotions or feedback. This results in a lack of consideration for the user in notifying them of the results, making it difficult to provide an optimal customer experience.

[0132] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0133] In this invention, the server includes means for acquiring multi-angle images of returned equipment, means for transmitting the acquired image data to a processing device, means for extracting damage features from the transmitted image data using analysis techniques, means for comparing the extracted features with past data to evaluate the damage, means for determining whether compensation is applicable based on the evaluation results, and means for optimizing the content of the transmission when communicating the judgment results, taking into account the user's emotional state. This enables accurate damage evaluation, allows for notifications that take the user's emotions into consideration, and provides an improved customer experience.

[0134] "Returned equipment" refers to electronic devices and equipment that have been returned after being used by a user.

[0135] "Multi-angle images" refer to multiple images of an object taken from different angles, providing three-dimensional information.

[0136] A "processing device" refers to a computer or server device that receives data and performs analysis and evaluation.

[0137] "Analysis techniques" refer to algorithms and methods for extracting meaningful information from data, and include image analysis and natural language processing.

[0138] "Damage characteristics" refer to specific data points or patterns that indicate physical defects or abnormalities that have occurred in the equipment.

[0139] "Past data" refers to stored information in a database that includes information about damage collected previously.

[0140] "Damage assessment" refers to the process of determining the degree and importance of equipment damage based on extracted characteristics.

[0141] "Eligible for compensation" refers to equipment that has been determined to be covered by warranty or repair, depending on the extent and type of damage.

[0142] "User's emotional state" refers to the emotional aspects of the feedback provided by the device user, including states such as dissatisfaction and satisfaction.

[0143] "Optimizing message content" refers to the process of adjusting the content of the message being conveyed, taking into account the emotional state of the user.

[0144] This invention provides a system that delivers more accurate and user-friendly notifications by combining damage assessment of returned devices with user emotion recognition. Specific embodiments are described below.

[0145] The terminal places the returned device in a dedicated shooting booth. This shooting booth is equipped with cameras that can photograph the device from multiple angles, thereby automatically acquiring multiple images. The images are saved in JPEG format and sent to a server via the network.

[0146] The server supplies the received image data to an AI image analysis module. The AI ​​technology used here leverages libraries such as TensorFlow and OpenCV to extract damage features from the images. The extracted data is compared with a historical database on the server. Machine learning algorithms are used for this comparison to determine whether compensation is necessary by calculating similarity.

[0147] Subsequently, the voice and text feedback data sent by the user is analyzed by an emotion recognition module. This analysis utilizes natural language processing (NLP) and voice feature extraction techniques. Specifically, it uses Google's Natural Language API to analyze the tone of emotion.

[0148] If a user's emotions are negative, the server takes this emotional state into consideration and optimizes the notification content. It utilizes a generative AI model to generate information in easy-to-understand language and provides a careful explanation to the user. This process enables not only accurate assessment of damage but also the provision of user-friendly services.

[0149] As a concrete example, an example of a prompt message is shown below.

[0150] "Use an AI model to recognize emotions and suggest how to adjust notifications if the user's feedback indicates dissatisfaction."

[0151] This invention makes it possible to provide users with higher quality and more emotionally sensitive services, contributing to an improved customer experience.

[0152] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0153] Step 1:

[0154] The terminal places the returned equipment in a shooting booth and uses its camera to capture images from multiple angles. The input is the equipment itself, and the output is a JPEG image file. This allows for detailed capture of the equipment's appearance.

[0155] Step 2:

[0156] The device transmits captured images to the server via the network. The input is a JPEG image file, and the output is the image data received by the server. This process is carried out using a communication protocol.

[0157] Step 3:

[0158] The server passes the received image to the AI ​​image analysis module. The input is JPEG image data, and the output is damage feature information. In the image analysis, features of the damaged area are extracted using a convolutional neural network (CNN) with TensorFlow or OpenCV.

[0159] Step 4:

[0160] The server compares the extracted damage features with those from a historical database. The input is the damage feature information, and the output is the compensation decision. A machine learning algorithm calculates the similarity of the feature vectors to determine whether compensation is appropriate.

[0161] Step 5:

[0162] Emotion recognition is performed based on user feedback. The input is voice or text data from the user, and the output is the result of emotion analysis. The server uses natural language processing technology to analyze the emotional tone of the text and perform emotion recognition.

[0163] Step 6:

[0164] The server adjusts the notification content based on the emotion recognition results. The input is the compensation decision result and the emotional state, and the output is a customized notification message. A generative AI model is used to generate explanations that are convincing to the user. This enables information delivery that is considerate of the user.

[0165] (Application Example 2)

[0166] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0167] In recent years, consumers expect not only a simple assessment of item damage during returns and exchanges, but also consideration of their feelings and a more satisfactory response. However, current systems struggle to adequately recognize user emotions and respond accordingly at each step of the return process. To improve this situation, there is a need to develop a system that effectively assesses item damage while taking user emotions into account.

[0168] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0169] In this invention, the server includes means for acquiring visual information of returned items, means for analyzing the acquired visual information and extracting characteristics of damage to the items, and means for recognizing the user's emotional state and optimizing the notification of the judgment results. This enables damage assessment and notification that takes the user's emotions into consideration.

[0170] "Returned items" refer to goods that have been sent back by consumers, either used or unused.

[0171] "Visual information" refers to image data acquired using cameras, scanners, etc., and related information based on that data.

[0172] "Analysis" refers to the process of identifying and understanding specific features and patterns based on acquired data.

[0173] "Damage characteristics" refer to the characteristics and morphological changes related to deterioration or damage that have occurred to an item.

[0174] "Damage information" refers to a collection of data on damage to items that have been registered in the past.

[0175] "Eligible for compensation" refers to the criteria for which compensation measures such as repair, replacement, or refund are available if damage or defects are found.

[0176] "Emotional state" refers to the results of measuring and evaluating the psychological and emotional responses exhibited by the user.

[0177] "Optimizing notifications" refers to the process of adjusting the content and method of notifications according to the user's needs and circumstances, and delivering information in the most effective and user-satisfying way.

[0178] The system that realizes this application integrates damage assessment of returned items with user sentiment recognition.

[0179] The server first acquires visual information about the returned item. To do this, it uses the terminal's camera to take images of the item from multiple angles. The captured image data is then transmitted to the server via the internet.

[0180] The server then uses an image analysis module to analyze the visual information and extract features of damage to the object. These image analysis modules include artificial intelligence models such as TensorFlow and OpenCV.

[0181] The server then compares the extracted damage characteristics with previously stored damage information. This comparison determines whether the item is eligible for compensation.

[0182] Furthermore, the server recognizes the user's emotional state from their voice or text. At this stage, emotion analysis is performed using IBM Watson® NLU or similar technologies. Once the emotional state is recognized, the server optimizes the notification content of the judgment result. This results in notifications that include careful explanations that take the user's emotions into consideration.

[0183] For example, if a user expresses feelings of being "very shocked" after returning a product, they will be notified of the appropriate compensation procedures along with a detailed assessment of the damage.

[0184] An example of a prompt for the generating AI model is: "Based on the damage assessment results observed on the camera lens, please advise on how to structure the notification content, taking into account the degree of surprise or dissatisfaction the user is experiencing."

[0185] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0186] Step 1:

[0187] The device takes images of the returned item. It captures visual information by photographing the item from multiple angles. The input is the item itself, and the output is the captured image file. The captured image is then sent to the next step.

[0188] Step 2:

[0189] The server receives the captured image files and sends them to the AI ​​image analysis module. Here, the input is the image data acquired in step 1, and the output is data with the damage features extracted. TensorFlow or OpenCV is used to analyze the damage in the image and capture its features.

[0190] Step 3:

[0191] The server compares the extracted damage characteristics with past damage information in the database. The input is the damage characteristic data obtained in step 2, and the output is the result of the compensation eligibility determination. Compensation is determined by comparing it with similar cases in the past database.

[0192] Step 4:

[0193] The server recognizes the user's emotional state through voice and text feedback. Here, the input is the voice or text feedback provided by the user, and the output is the analyzed emotional state. IBM Watson NLU and similar systems are used to classify the user's emotions.

[0194] Step 5:

[0195] The server optimizes the notification content based on the damage compensation decision and the user's emotional state. The input is the results of steps 3 and 4, and the output is a customized notification message. The notification message is crafted to be sensitive to the user's emotions and is sent in the next step.

[0196] Step 6:

[0197] The server sends the optimized notification content to the user. Here, the input is the notification text generated in step 5, and the output is the notification the user receives. The notification is sent via email or a dedicated application, allowing the user to check it.

[0198] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0199] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0200] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0201] [Second Embodiment]

[0202] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0203] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0204] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0205] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0206] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0207] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0208] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0209] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0210] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0211] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0212] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0213] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0214] The image analysis system of the present invention is built to automate the process of efficiently evaluating returned equipment and analyzing its damage. Specific embodiments of this system are described below.

[0215] Upon arrival at the office, the returned device is placed in a dedicated shooting booth. This booth is equipped with a high-performance camera that captures images of the device from multiple angles. These images are immediately transferred to a server, and the subsequent analysis process begins.

[0216] The server receives the transmitted images and performs analysis within its AI image analysis module. This module utilizes deep learning technology to automatically extract damage features such as cracks and surface scratches from the images. This sophisticated process allows for accurate detection of even minute damage.

[0217] Next, the server compares the feature extraction results with the database. The database contains a collection of past damage data and corresponding reports, and uses this to calculate the similarity of the damage. If the similarity exceeds a certain threshold, the server determines that the damage is eligible for compensation.

[0218] Finally, the server notifies the user of the decision. This notification is sent via email or a dedicated application, and a detailed analysis report is attached to the result. Based on this report, the user can then initiate the compensation process.

[0219] As a concrete example, consider a case where a device is returned and a large crack is found on its screen. The server compares the shape of the crack with its similarity to past data and determines that it is human-caused damage. It then automatically calculates the compensation amount and notifies the user. This entire process ensures operational efficiency and fair judgment.

[0220] The following describes the processing flow.

[0221] Step 1:

[0222] The terminal is placed inside the shooting booth, and a high-resolution camera captures images from multiple angles of the exterior. The captured images are then transmitted directly to the server in digital format.

[0223] Step 2:

[0224] The server inputs the received image into the AI ​​image analysis module and starts the image analysis process. In this process, the server uses deep learning technology to extract damage features such as scratches and cracks.

[0225] Step 3:

[0226] The server compares the extracted features with historical damage data stored in its internal database. Here, it calculates the similarity between the current features and similar damage patterns reported in the past.

[0227] Step 4:

[0228] The server determines whether the damage is covered by compensation based on the similarity calculation results. This determination applies pre-defined criteria to determine whether compensation is necessary depending on the type and extent of the damage.

[0229] Step 5:

[0230] The server notifies the user of the decision. The notification is sent via email or a dedicated application, and a detailed analysis report is attached. This allows the user to immediately initiate the necessary compensation procedures.

[0231] (Example 1)

[0232] Next, we will describe Example 1. 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".

[0233] In assessing damage to returned equipment, conventional methods often rely heavily on human judgment, limiting the consistency and speed of evaluations. Furthermore, there is a need for highly accurate analysis of diverse damage patterns. Additionally, there is a lack of effective means to communicate damage assessment results to users, highlighting the need for increased efficiency and fairness in decision-making.

[0234] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0235] In this invention, the server includes means for installing returned equipment and acquiring images from multiple angles in a controlled environment; means for automatically extracting damage features of the equipment using a deep learning algorithm; and means for comparing the acquired damage features with past data and evaluating the similarity. This enables accurate and rapid damage assessment and fair compensation decisions while minimizing human intervention.

[0236] "Returned equipment" refers to devices that have been used by a user and then returned to their original location.

[0237] A "controlled shooting environment" refers to a shooting booth configuration where the light source and shooting conditions are standardized, minimizing the influence of the external environment.

[0238] A "deep learning algorithm" refers to a technology that automatically learns specific features from images through training with large amounts of data, and performs damage detection and classification.

[0239] "Damage characteristics" refer to information about the location, shape, and size of cracks and damage extracted through image analysis.

[0240] "Similarity assessment" refers to the process of comparing extracted damage features with data in past databases and quantifying their correlation.

[0241] A "communication network" refers to a network infrastructure built for the purpose of exchanging data.

[0242] "User" refers to the final recipient of a system or service.

[0243] This invention provides a system for automatically evaluating the damage to returned equipment and efficiently determining whether compensation is applicable. A specific embodiment of this system is described below.

[0244] First, when a user returns a device to the office, it is placed in a dedicated shooting booth. The shooting booth provides a controlled environment, maintaining a constant position for the light source and subject, making it possible to acquire high-definition images every time, eliminating the influence of ambient light. This shooting booth is equipped with a high-performance digital camera that captures detailed images from multiple angles of the device.

[0245] Next, the server quickly receives the captured images and stores them in a central storage device. This data is immediately sent to an AI analysis module, where it is automatically processed using a deep learning algorithm. This algorithm, built using TensorFlow, automatically extracts damage features and recognizes crack and damage patterns with high accuracy. This allows users to quickly detect even minute damage that might be overlooked during human visual inspection.

[0246] Subsequently, the server compares the extracted damage characteristics with past damage information stored in a database. This database records past damage cases, their causes, and repair history, and the server evaluates the similarity to this information to determine compensation. This evaluation result is standardized through an automated process, enabling fair and consistent decisions.

[0247] Ultimately, the server determines whether compensation is applicable and notifies the user of the result. This notification is sent via email or a dedicated application, allowing the user to check the result immediately. The notification includes an analysis report detailing the damage, which the user can use to proceed with the subsequent procedures.

[0248] For example, if a user returns a device with a cracked screen, the server analyzes the shape of the crack by comparing it to past data and determines that it is human-caused damage. At this point, the server automatically calculates the compensation amount and notifies the user. This process allows users to receive compensation quickly and fairly.

[0249] An example of a prompt when using a generative AI model is: "Please analyze the damage to the returned device in detail and inform us about the possibility of compensation."

[0250] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0251] Step 1:

[0252] Upon returning the device, the user places it in a dedicated shooting booth. This booth eliminates ambient light and uses a high-performance camera to capture images of the device from multiple angles. The input data consists of the captured high-resolution images of the device. This results in clear image data being obtained as output.

[0253] Step 2:

[0254] The server receives captured images via the network and stores them in a central repository. The input consists of multiple image files received by the server. The server organizes these image data appropriately and prepares them for immediate use in the subsequent analysis process. The output of this step is the image data ready for analysis.

[0255] Step 3:

[0256] The server passes the saved image to the AI ​​image analysis module and starts the analysis. The input is image data ready for analysis. The server uses a deep learning algorithm powered by TensorFlow to automatically extract damage features such as cracks and scratches from the image. The output includes the extracted damage features as data.

[0257] Step 4:

[0258] The server compares the extracted damage features with a database of past damages and calculates similarity. The inputs are the extracted damage features and the database of past damages. The server uses statistical methods to evaluate the similarity and establish criteria for determining whether compensation is necessary. The output of this step is the damage similarity score.

[0259] Step 5:

[0260] The server makes a compensation decision based on the similarity score and notifies the user of the result. The input is the damage similarity score. The server generates the result and reports it to the user via email or a dedicated application. The output is a notification of the decision result, including a detailed analysis report.

[0261] (Application Example 1)

[0262] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0263] In the damage assessment process for returned goods at logistics centers, there is a need to reduce the variability of human judgment and automate the process for quick and fair compensation decisions. Furthermore, building a system that improves work efficiency by providing real-time notification of results is a key challenge.

[0264] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0265] In this invention, the server includes means for acquiring images of returned items, means for analyzing the acquired images and extracting characteristics of the damage to the items, means for comparing the extracted characteristics with previously registered damage data, means for determining whether the items are eligible for compensation based on the results of the comparison, means for notifying the determination result, and means for the notification means to provide the result in real time via a mobile terminal device. This makes it possible to automate the damage assessment of returned goods at a logistics center and to quickly make fair compensation decisions.

[0266] "Returned items" refers to all products and goods that have been sent back after being used by a customer or user.

[0267] "Means of acquiring images" refers to devices and technologies for visually capturing the whole or parts of an object and saving them as digital data.

[0268] "Means for analyzing images and extracting the characteristics of damage to an item" refers to technologies and devices that process acquired image data to identify damage to the surface or interior of an item and automatically recognize its detailed characteristics.

[0269] "Means of comparing with previously registered damage data" refers to technologies and processes for matching information on previously recorded damage with newly identified damage characteristics.

[0270] "Means of determining whether an item is eligible for compensation" refers to criteria or devices used to determine whether an item meets the compensation requirements, based on the characteristics of the damage and past data.

[0271] "Means of notifying the decision result" refers to communication technologies and systems used to convey information to relevant parties regarding whether compensation is possible and details of the damage.

[0272] "Means of providing results in real time via mobile terminal devices" refers to technologies and systems that allow users to receive decision results immediately through portable digital devices.

[0273] The system implementing this invention incorporates advanced image analysis technology to efficiently assess the damage to returned items.

[0274] First, the items returned by the user are placed in a dedicated shooting environment, and multiple images are acquired from the entire object and from the necessary angles using the camera of a mobile device. This mobile device has a dedicated application installed for transferring the images to a server in the cloud.

[0275] The server performs AI image analysis using the received image data. Specifically, a model utilizing deep learning technology automatically extracts damage features and compares this information with past damage data stored in a database. The AI ​​model employs machine learning techniques using TensorFlow.

[0276] The server determines the need for compensation based on the matching results. If compensation is deemed necessary, the server notifies the user via the mobile device. The notification includes details of the damage and information regarding the application of compensation. The results are provided in real time using the push notification technology used.

[0277] As a concrete example, consider a case where a household electrical appliance returned to a logistics center has a cracked screen due to a fall. This system identifies the damage in real time and immediately determines whether it is eligible for compensation by comparing it with past data.

[0278] An example of a prompt message might be: "From this image, detect the extent of damage to the product upon return and report it by comparing it to similar data." This inquiry is then posed to the generating AI model.

[0279] This will not only dramatically improve operational efficiency at logistics centers, but also ensure fairness and speed in compensation.

[0280] The process of a specific process in Application Example 1 will be described using FIG. 12.

[0281] Step 1:

[0282] The user takes a picture of the returned item using the camera of the mobile terminal device. The user takes pictures of the entire item and the suspected damaged areas from multiple angles and inputs the image data through the dedicated application of the terminal. The input image data is transferred to the server on the cloud.

[0283] Step 2:

[0284] The server passes the received image data to the AI analysis module. The server uses a generative AI model using TensorFlow to analyze the damage characteristics of the item from the image. Here, information extracted as damage characteristics is output for the input of the image data. As a specific operation, data operations are performed to detect crack and deformation patterns.

[0285] Step 3:

[0286] The server compares the extracted damage characteristics with the past damage data stored in the database. The server receives the damage information of the analysis result as input and calculates the similarity with past compensation cases. As a result, a determination result as to whether the damage is a compensation target is output. As a specific operation, data comparison is performed using a similarity calculation algorithm.

[0287] Step 4:

[0288] The server notifies the mobile terminal device of the determination result and conveys it to the user. The notification from the server includes the detailed information of the damage and the determination result regarding compensation and is pushed to the user in real time. The output of the notification serves as a basis for the user to make a compensation application. As a specific operation, the result is transmitted via the notification API.

[0289] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0290] This invention provides a system that combines user emotion recognition with a device damage assessment process. This system not only analyzes the damage to returned devices using image analysis and makes compensation decisions based on the analysis results, but also recognizes the user's emotional state and optimizes the notification process.

[0291] First, the returned device is placed in a shooting booth, and images are automatically taken from multiple angles. The captured images are immediately sent to a server. This image data is analyzed by an AI image analysis module, and damage characteristics are extracted. The server compares the extracted characteristics with an internal historical database to determine whether the damage is eligible for compensation.

[0292] Subsequently, the server considers the user's emotions when notifying them of the decision. This function is provided by an emotion recognition module, which analyzes the user's voice and written feedback data. If the user's emotions are negative, such as dissatisfaction or surprise, the server adjusts the notification content to include a polite and convincing explanation before sending it.

[0293] For example, if a user is dissatisfied with device damage, emotion recognition can detect that emotion. In this case, the server sends a report in addition to the usual notification, which includes a detailed explanation of the compensation and information on how the damage was assessed. This demonstrates consideration for the user and improves the customer experience.

[0294] This invention makes it possible to provide higher quality service not only by accurately assessing damage, but also by incorporating user emotional responses.

[0295] The following describes the processing flow.

[0296] Step 1:

[0297] Once the device is returned, it is placed in the shooting booth. A high-resolution camera captures images of various parts of the device from multiple angles, and the acquired images are sent to a server as digital data.

[0298] Step 2:

[0299] The server inputs the received image data into an AI image analysis module. This module extracts damage features such as scratches and cracks from the image. The extracted features are detailed data indicating the type, location, and extent of the damage.

[0300] Step 3:

[0301] The server compares the extracted damage characteristics with a historical database and calculates the similarity. The database contains previously recorded damage cases, and the server uses this comparison to determine whether the damage is eligible for compensation.

[0302] Step 4:

[0303] Based on the decision, the server creates notification content while considering the user's emotional state. The emotion recognition module analyzes the voice data and feedback obtained from the user to determine the user's emotions.

[0304] Step 5:

[0305] The server responds flexibly to the user's emotional state. For example, if negative emotions are detected, the notification will include additional information about compensation details and the decision-making process, demonstrating consideration for the user.

[0306] Step 6:

[0307] Finally, the server sends the judgment result and the adjusted notification content to the user. As a result, the user can accurately understand the result of the damage assessment and the compensation content, and can move on to the necessary procedures.

[0308] (Example 2)

[0309] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0310] Conventional damage assessment systems have a problem in that when making a damage judgment on a returned device, they only rely on objective data regarding the state of the device, and thus cannot take into account the user's feelings and feedback. Therefore, there is a lack of consideration for the user in the result notification, and it is difficult to provide an optimal customer experience.

[0311] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0312] In this invention, the server includes means for acquiring multi-angle images of the returned device, means for transmitting the acquired image data to the processing device, means for extracting damage features from the transmitted image data using analysis techniques, means for comparing the extracted features with past data to perform damage assessment, means for determining whether it is a compensation target based on the assessment result, and means for optimizing the transmitted content by considering the user's emotional state when transmitting the determination result. As a result, accurate damage assessment can be realized, notification considering the user's feelings becomes possible, and it becomes possible to provide an improved customer experience.

[0313] The "returned device" refers to an electronic device or apparatus that has been returned after being used by the user.

[0314] The "multi-angle image" refers to a plurality of images taken of an object from different directions, providing three-dimensional information.

[0315] A "processing device" refers to a computer or server device that receives data and performs analysis and evaluation.

[0316] "Analysis techniques" refer to algorithms and methods for extracting meaningful information from data, and include image analysis and natural language processing.

[0317] "Damage characteristics" refer to specific data points or patterns that indicate physical defects or abnormalities that have occurred in the equipment.

[0318] "Past data" refers to stored information in a database that includes information about damage collected previously.

[0319] "Damage assessment" refers to the process of determining the degree and importance of equipment damage based on extracted characteristics.

[0320] "Eligible for compensation" refers to equipment that has been determined to be covered by warranty or repair, depending on the extent and type of damage.

[0321] "User's emotional state" refers to the emotional aspects of the feedback provided by the device user, including states such as dissatisfaction and satisfaction.

[0322] "Optimizing message content" refers to the process of adjusting the content of the message being conveyed, taking into account the emotional state of the user.

[0323] This invention provides a system that delivers more accurate and user-friendly notifications by combining damage assessment of returned devices with user emotion recognition. Specific embodiments are described below.

[0324] The terminal places the returned device in a dedicated shooting booth. This shooting booth is equipped with cameras that can photograph the device from multiple angles, thereby automatically acquiring multiple images. The images are saved in JPEG format and sent to a server via the network.

[0325] The server supplies the received image data to an AI image analysis module. The AI ​​technology used here leverages libraries such as TensorFlow and OpenCV to extract damage features from the images. The extracted data is compared with a historical database on the server. Machine learning algorithms are used for this comparison to determine whether compensation is necessary by calculating similarity.

[0326] Subsequently, the voice and text feedback data sent by the user is analyzed by an emotion recognition module. This analysis utilizes natural language processing (NLP) and voice feature extraction techniques. Specifically, it uses Google's Natural Language API to analyze the tone of emotion.

[0327] If a user's emotions are negative, the server takes this emotional state into consideration and optimizes the notification content. It utilizes a generative AI model to generate information in easy-to-understand language and provides a careful explanation to the user. This process enables not only accurate assessment of damage but also the provision of user-friendly services.

[0328] As a concrete example, an example of a prompt message is shown below.

[0329] "Use an AI model to recognize emotions and suggest how to adjust notifications if the user's feedback indicates dissatisfaction."

[0330] This invention makes it possible to provide users with higher quality and more emotionally sensitive services, contributing to an improved customer experience.

[0331] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0332] Step 1:

[0333] The terminal places the returned equipment in a shooting booth and uses its camera to capture images from multiple angles. The input is the equipment itself, and the output is a JPEG image file. This allows for detailed capture of the equipment's appearance.

[0334] Step 2:

[0335] The device transmits captured images to the server via the network. The input is a JPEG image file, and the output is the image data received by the server. This process is carried out using a communication protocol.

[0336] Step 3:

[0337] The server passes the received image to the AI ​​image analysis module. The input is JPEG image data, and the output is damage feature information. In the image analysis, features of the damaged area are extracted using a convolutional neural network (CNN) with TensorFlow or OpenCV.

[0338] Step 4:

[0339] The server compares the extracted damage features with those from a historical database. The input is the damage feature information, and the output is the compensation decision. A machine learning algorithm calculates the similarity of the feature vectors to determine whether compensation is appropriate.

[0340] Step 5:

[0341] Emotion recognition is performed based on user feedback. The input is voice or text data from the user, and the output is the result of emotion analysis. The server uses natural language processing technology to analyze the emotional tone of the text and perform emotion recognition.

[0342] Step 6:

[0343] The server adjusts the notification content based on the emotion recognition results. The input is the compensation decision result and the emotional state, and the output is a customized notification message. A generative AI model is used to generate explanations that are convincing to the user. This enables information delivery that is considerate of the user.

[0344] (Application Example 2)

[0345] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0346] In recent years, consumers expect not only a simple assessment of item damage during returns and exchanges, but also consideration of their feelings and a more satisfactory response. However, current systems struggle to adequately recognize user emotions and respond accordingly at each step of the return process. To improve this situation, there is a need to develop a system that effectively assesses item damage while taking user emotions into account.

[0347] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0348] In this invention, the server includes means for acquiring visual information of returned items, means for analyzing the acquired visual information and extracting characteristics of damage to the items, and means for recognizing the user's emotional state and optimizing the notification of the judgment results. This enables damage assessment and notification that takes the user's emotions into consideration.

[0349] "Returned items" refer to goods that have been sent back by consumers, either used or unused.

[0350] "Visual information" refers to image data acquired using cameras, scanners, etc., and related information based on that data.

[0351] "Analysis" refers to the process of identifying and understanding specific features and patterns based on acquired data.

[0352] "Damage characteristics" refer to the characteristics and morphological changes related to deterioration or damage that have occurred to an item.

[0353] "Damage information" refers to a collection of data on damage to items that have been registered in the past.

[0354] "Eligible for compensation" refers to the criteria for which compensation measures such as repair, replacement, or refund are available if damage or defects are found.

[0355] "Emotional state" refers to the results of measuring and evaluating the psychological and emotional responses exhibited by the user.

[0356] "Optimizing notifications" refers to the process of adjusting the content and method of notifications according to the user's needs and circumstances, and delivering information in the most effective and user-satisfying way.

[0357] The system that realizes this application integrates damage assessment of returned items with user sentiment recognition.

[0358] The server first acquires visual information about the returned item. To do this, it uses the terminal's camera to take images of the item from multiple angles. The captured image data is then transmitted to the server via the internet.

[0359] The server then uses an image analysis module to analyze the visual information and extract features of damage to the object. These image analysis modules include artificial intelligence models such as TensorFlow and OpenCV.

[0360] The server then compares the extracted damage characteristics with previously stored damage information. This comparison determines whether the item is eligible for compensation.

[0361] Furthermore, the server recognizes the user's emotional state from their voice or text. At this stage, emotion analysis is performed using tools such as IBM Watson NLU. Once the emotional state is recognized, the server optimizes the notification content of the judgment result. This results in notifications that include careful explanations that take the user's emotions into consideration.

[0362] For example, if a user expresses feelings of being "very shocked" after returning a product, they will be notified of the appropriate compensation procedures along with a detailed assessment of the damage.

[0363] An example of a prompt for the generating AI model is: "Based on the damage assessment results observed on the camera lens, please advise on how to structure the notification content, taking into account the degree of surprise or dissatisfaction the user is experiencing."

[0364] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0365] Step 1:

[0366] The device takes images of the returned item. It captures visual information by photographing the item from multiple angles. The input is the item itself, and the output is the captured image file. The captured image is then sent to the next step.

[0367] Step 2:

[0368] The server receives the captured image files and sends them to the AI ​​image analysis module. Here, the input is the image data acquired in step 1, and the output is data with the damage features extracted. TensorFlow or OpenCV is used to analyze the damage in the image and capture its features.

[0369] Step 3:

[0370] The server compares the extracted damage characteristics with past damage information in the database. The input is the damage characteristic data obtained in step 2, and the output is the result of the compensation eligibility determination. Compensation is determined by comparing it with similar cases in the past database.

[0371] Step 4:

[0372] The server recognizes the user's emotional state through voice and text feedback. Here, the input is the voice or text feedback provided by the user, and the output is the analyzed emotional state. IBM Watson NLU and similar systems are used to classify the user's emotions.

[0373] Step 5:

[0374] The server optimizes the notification content based on the damage compensation decision and the user's emotional state. The input is the results of steps 3 and 4, and the output is a customized notification message. The notification message is crafted to be sensitive to the user's emotions and is sent in the next step.

[0375] Step 6:

[0376] The server sends the optimized notification content to the user. Here, the input is the notification text generated in step 5, and the output is the notification the user receives. The notification is sent via email or a dedicated application, allowing the user to check it.

[0377] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0378] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0379] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0380] [Third Embodiment]

[0381] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0382] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0383] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0384] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0385] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0386] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0387] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0388] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0389] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0390] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0391] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0392] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0393] The image analysis system of the present invention is built to automate the process of efficiently evaluating returned equipment and analyzing its damage. Specific embodiments of this system are described below.

[0394] Upon arrival at the office, the returned device is placed in a dedicated shooting booth. This booth is equipped with a high-performance camera that captures images of the device from multiple angles. These images are immediately transferred to a server, and the subsequent analysis process begins.

[0395] The server receives the transmitted images and performs analysis within its AI image analysis module. This module utilizes deep learning technology to automatically extract damage features such as cracks and surface scratches from the images. This sophisticated process allows for accurate detection of even minute damage.

[0396] Next, the server compares the feature extraction results with the database. The database contains a collection of past damage data and corresponding reports, and uses this to calculate the similarity of the damage. If the similarity exceeds a certain threshold, the server determines that the damage is eligible for compensation.

[0397] Finally, the server notifies the user of the decision. This notification is sent via email or a dedicated application, and a detailed analysis report is attached to the result. Based on this report, the user can then initiate the compensation process.

[0398] As a concrete example, consider a case where a device is returned and a large crack is found on its screen. The server compares the shape of the crack with its similarity to past data and determines that it is human-caused damage. It then automatically calculates the compensation amount and notifies the user. This entire process ensures operational efficiency and fair judgment.

[0399] The following describes the processing flow.

[0400] Step 1:

[0401] The terminal is placed inside the shooting booth, and a high-resolution camera captures images from multiple angles of the exterior. The captured images are then transmitted directly to the server in digital format.

[0402] Step 2:

[0403] The server inputs the received image into the AI ​​image analysis module and starts the image analysis process. In this process, the server uses deep learning technology to extract damage features such as scratches and cracks.

[0404] Step 3:

[0405] The server compares the extracted features with historical damage data stored in its internal database. Here, it calculates the similarity between the current features and similar damage patterns reported in the past.

[0406] Step 4:

[0407] The server determines whether the damage is covered by compensation based on the similarity calculation results. This determination applies pre-defined criteria to determine whether compensation is necessary depending on the type and extent of the damage.

[0408] Step 5:

[0409] The server notifies the user of the decision. The notification is sent via email or a dedicated application, and a detailed analysis report is attached. This allows the user to immediately initiate the necessary compensation procedures.

[0410] (Example 1)

[0411] Next, we will describe Example 1. 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."

[0412] In assessing damage to returned equipment, conventional methods often rely heavily on human judgment, limiting the consistency and speed of evaluations. Furthermore, there is a need for highly accurate analysis of diverse damage patterns. Additionally, there is a lack of effective means to communicate damage assessment results to users, highlighting the need for increased efficiency and fairness in decision-making.

[0413] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0414] In this invention, the server includes means for installing returned equipment and acquiring images from multiple angles in a controlled environment; means for automatically extracting damage features of the equipment using a deep learning algorithm; and means for comparing the acquired damage features with past data and evaluating the similarity. This enables accurate and rapid damage assessment and fair compensation decisions while minimizing human intervention.

[0415] "Returned equipment" refers to devices that have been used by a user and then returned to their original location.

[0416] A "controlled shooting environment" refers to a shooting booth configuration where the light source and shooting conditions are standardized, minimizing the influence of the external environment.

[0417] A "deep learning algorithm" refers to a technology that automatically learns specific features from images through training with large amounts of data, and performs damage detection and classification.

[0418] "Damage characteristics" refer to information about the location, shape, and size of cracks and damage extracted through image analysis.

[0419] "Similarity assessment" refers to the process of comparing extracted damage features with data in past databases and quantifying their correlation.

[0420] A "communication network" refers to a network infrastructure built for the purpose of exchanging data.

[0421] "User" refers to the final recipient of a system or service.

[0422] This invention provides a system for automatically evaluating the damage to returned equipment and efficiently determining whether compensation is applicable. A specific embodiment of this system is described below.

[0423] First, when a user returns a device to the office, it is placed in a dedicated shooting booth. The shooting booth provides a controlled environment, maintaining a constant position for the light source and subject, making it possible to acquire high-definition images every time, eliminating the influence of ambient light. This shooting booth is equipped with a high-performance digital camera that captures detailed images from multiple angles of the device.

[0424] Next, the server quickly receives the captured images and stores them in a central storage device. This data is immediately sent to an AI analysis module, where it is automatically processed using a deep learning algorithm. This algorithm, built using TensorFlow, automatically extracts damage features and recognizes crack and damage patterns with high accuracy. This allows users to quickly detect even minute damage that might be overlooked during human visual inspection.

[0425] Subsequently, the server compares the extracted damage characteristics with past damage information stored in a database. This database records past damage cases, their causes, and repair history, and the server evaluates the similarity to this information to determine compensation. This evaluation result is standardized through an automated process, enabling fair and consistent decisions.

[0426] Ultimately, the server determines whether compensation is applicable and notifies the user of the result. This notification is sent via email or a dedicated application, allowing the user to check the result immediately. The notification includes an analysis report detailing the damage, which the user can use to proceed with the subsequent procedures.

[0427] For example, if a user returns a device with a cracked screen, the server analyzes the shape of the crack by comparing it to past data and determines that it is human-caused damage. At this point, the server automatically calculates the compensation amount and notifies the user. This process allows users to receive compensation quickly and fairly.

[0428] An example of a prompt when using a generative AI model is: "Please analyze the damage to the returned device in detail and inform us about the possibility of compensation."

[0429] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0430] Step 1:

[0431] Upon returning the device, the user places it in a dedicated shooting booth. This booth eliminates ambient light and uses a high-performance camera to capture images of the device from multiple angles. The input data consists of the captured high-resolution images of the device. This results in clear image data being obtained as output.

[0432] Step 2:

[0433] The server receives captured images via the network and stores them in a central repository. The input consists of multiple image files received by the server. The server organizes these image data appropriately and prepares them for immediate use in the subsequent analysis process. The output of this step is the image data ready for analysis.

[0434] Step 3:

[0435] The server passes the saved image to the AI ​​image analysis module and starts the analysis. The input is image data ready for analysis. The server uses a deep learning algorithm powered by TensorFlow to automatically extract damage features such as cracks and scratches from the image. The output includes the extracted damage features as data.

[0436] Step 4:

[0437] The server compares the extracted damage features with a database of past damages and calculates similarity. The inputs are the extracted damage features and the database of past damages. The server uses statistical methods to evaluate the similarity and establish criteria for determining whether compensation is necessary. The output of this step is the damage similarity score.

[0438] Step 5:

[0439] The server makes a compensation decision based on the similarity score and notifies the user of the result. The input is the damage similarity score. The server generates the result and reports it to the user via email or a dedicated application. The output is a notification of the decision result, including a detailed analysis report.

[0440] (Application Example 1)

[0441] Next, we will explain Application Example 1. In the following explanation, 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."

[0442] In the damage assessment process for returned goods at logistics centers, there is a need to reduce the variability of human judgment and automate the process for quick and fair compensation decisions. Furthermore, building a system that improves work efficiency by providing real-time notification of results is a key challenge.

[0443] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0444] In this invention, the server includes means for acquiring images of returned items, means for analyzing the acquired images and extracting characteristics of the damage to the items, means for comparing the extracted characteristics with previously registered damage data, means for determining whether the items are eligible for compensation based on the results of the comparison, means for notifying the determination result, and means for the notification means to provide the result in real time via a mobile terminal device. This makes it possible to automate the damage assessment of returned goods at a logistics center and to quickly make fair compensation decisions.

[0445] "Returned items" refers to all products and goods that have been sent back after being used by a customer or user.

[0446] "Means of acquiring images" refers to devices and technologies for visually capturing the whole or parts of an object and saving them as digital data.

[0447] "Means for analyzing images and extracting the characteristics of damage to an item" refers to technologies and devices that process acquired image data to identify damage to the surface or interior of an item and automatically recognize its detailed characteristics.

[0448] "Means of comparing with previously registered damage data" refers to technologies and processes for matching information on previously recorded damage with newly identified damage characteristics.

[0449] "Means of determining whether an item is eligible for compensation" refers to criteria or devices used to determine whether an item meets the compensation requirements, based on the characteristics of the damage and past data.

[0450] "Means of notifying the decision result" refers to communication technologies and systems used to convey information to relevant parties regarding whether compensation is possible and details of the damage.

[0451] "Means of providing results in real time via mobile terminal devices" refers to technologies and systems that allow users to receive decision results immediately through portable digital devices.

[0452] The system implementing this invention incorporates advanced image analysis technology to efficiently assess the damage to returned items.

[0453] First, the items returned by the user are placed in a dedicated shooting environment, and multiple images are acquired from the entire object and from the necessary angles using the camera of a mobile device. This mobile device has a dedicated application installed for transferring the images to a server in the cloud.

[0454] The server performs AI image analysis using the received image data. Specifically, a model utilizing deep learning technology automatically extracts damage features and compares this information with past damage data stored in a database. The AI ​​model employs machine learning techniques using TensorFlow.

[0455] The server determines the need for compensation based on the matching results. If compensation is deemed necessary, the server notifies the user via the mobile device. The notification includes details of the damage and information regarding the application of compensation. The results are provided in real time using the push notification technology used.

[0456] As a concrete example, consider a case where a household electrical appliance returned to a logistics center has a cracked screen due to a fall. This system identifies the damage in real time and immediately determines whether it is eligible for compensation by comparing it with past data.

[0457] An example of a prompt message might be: "From this image, detect the extent of damage to the product upon return and report it by comparing it to similar data." This inquiry is then posed to the generating AI model.

[0458] This will not only dramatically improve operational efficiency at logistics centers, but also ensure fairness and speed in compensation.

[0459] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0460] Step 1:

[0461] The user photographs the returned item using the camera on their mobile device. The user photographs the entire item and any suspected damaged areas from multiple angles and inputs the image data through a dedicated application on the device. The input image data is then transferred to a server in the cloud.

[0462] Step 2:

[0463] The server passes the received image data to the AI ​​analysis module. The server uses a generative AI model based on TensorFlow to analyze the damage features of the object from the image. Here, information extracted as damage features is output in response to the image data input. Specifically, it performs data calculations to detect crack and deformation patterns.

[0464] Step 3:

[0465] The server compares the extracted damage features with historical damage data stored in the database. The server receives the damage information from the analysis as input and calculates its similarity to past compensation cases. This results in an output determining whether the damage is eligible for compensation. Specifically, it performs data matching using a similarity calculation algorithm.

[0466] Step 4:

[0467] The server notifies the mobile device of the assessment result and communicates it to the user. The notification from the server includes detailed information about the damage and the assessment result regarding compensation, and is pushed to the user in real time. The output of the notification serves as a basis for the user to decide whether to file a compensation claim. Specifically, the result is sent via a notification API.

[0468] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0469] This invention provides a system that combines user emotion recognition with a device damage assessment process. This system not only analyzes the damage to returned devices using image analysis and makes compensation decisions based on the analysis results, but also recognizes the user's emotional state and optimizes the notification process.

[0470] First, the returned device is placed in a shooting booth, and images are automatically taken from multiple angles. The captured images are immediately sent to a server. This image data is analyzed by an AI image analysis module, and damage characteristics are extracted. The server compares the extracted characteristics with an internal historical database to determine whether the damage is eligible for compensation.

[0471] Subsequently, the server considers the user's emotions when notifying them of the decision. This function is provided by an emotion recognition module, which analyzes the user's voice and written feedback data. If the user's emotions are negative, such as dissatisfaction or surprise, the server adjusts the notification content to include a polite and convincing explanation before sending it.

[0472] For example, if a user is dissatisfied with device damage, emotion recognition can detect that emotion. In this case, the server sends a report in addition to the usual notification, which includes a detailed explanation of the compensation and information on how the damage was assessed. This demonstrates consideration for the user and improves the customer experience.

[0473] This invention makes it possible to provide higher quality service not only by accurately assessing damage, but also by incorporating user emotional responses.

[0474] The following describes the processing flow.

[0475] Step 1:

[0476] Once the device is returned, it is placed in the shooting booth. A high-resolution camera captures images of various parts of the device from multiple angles, and the acquired images are sent to a server as digital data.

[0477] Step 2:

[0478] The server inputs the received image data into an AI image analysis module. This module extracts damage features such as scratches and cracks from the image. The extracted features are detailed data indicating the type, location, and extent of the damage.

[0479] Step 3:

[0480] The server compares the extracted damage characteristics with a historical database and calculates the similarity. The database contains previously recorded damage cases, and the server uses this comparison to determine whether the damage is eligible for compensation.

[0481] Step 4:

[0482] Based on the decision, the server creates notification content while considering the user's emotional state. The emotion recognition module analyzes the voice data and feedback obtained from the user to determine the user's emotions.

[0483] Step 5:

[0484] The server responds flexibly to the user's emotional state. For example, if negative emotions are detected, the notification will include additional information about compensation details and the decision-making process, demonstrating consideration for the user.

[0485] Step 6:

[0486] Finally, the server sends the assessment results and adjusted notification to the user. This allows the user to accurately understand the damage assessment results and compensation details and proceed with the necessary procedures.

[0487] (Example 2)

[0488] Next, we will describe Example 2. 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."

[0489] Conventional damage assessment systems rely solely on objective data regarding the condition of returned equipment when determining damage, which means they cannot take into account user emotions or feedback. This results in a lack of consideration for the user in notifying them of the results, making it difficult to provide an optimal customer experience.

[0490] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0491] In this invention, the server includes means for acquiring multi-angle images of returned equipment, means for transmitting the acquired image data to a processing device, means for extracting damage features from the transmitted image data using analysis techniques, means for comparing the extracted features with past data to evaluate the damage, means for determining whether compensation is applicable based on the evaluation results, and means for optimizing the content of the transmission when communicating the judgment results, taking into account the user's emotional state. This enables accurate damage evaluation, allows for notifications that take the user's emotions into consideration, and provides an improved customer experience.

[0492] "Returned equipment" refers to electronic devices and equipment that have been returned after being used by a user.

[0493] "Multi-angle images" refer to multiple images of an object taken from different angles, providing three-dimensional information.

[0494] A "processing device" refers to a computer or server device that receives data and performs analysis and evaluation.

[0495] "Analysis techniques" refer to algorithms and methods for extracting meaningful information from data, and include image analysis and natural language processing.

[0496] "Damage characteristics" refer to specific data points or patterns that indicate physical defects or abnormalities that have occurred in the equipment.

[0497] "Past data" refers to stored information in a database that includes information about damage collected previously.

[0498] "Damage assessment" refers to the process of determining the degree and importance of equipment damage based on extracted characteristics.

[0499] "Eligible for compensation" refers to equipment that has been determined to be covered by warranty or repair, depending on the extent and type of damage.

[0500] "User's emotional state" refers to the emotional aspects of the feedback provided by the device user, including states such as dissatisfaction and satisfaction.

[0501] "Optimizing message content" refers to the process of adjusting the content of the message being conveyed, taking into account the emotional state of the user.

[0502] This invention provides a system that delivers more accurate and user-friendly notifications by combining damage assessment of returned devices with user emotion recognition. Specific embodiments are described below.

[0503] The terminal places the returned device in a dedicated shooting booth. This shooting booth is equipped with cameras that can photograph the device from multiple angles, thereby automatically acquiring multiple images. The images are saved in JPEG format and sent to a server via the network.

[0504] The server supplies the received image data to an AI image analysis module. The AI ​​technology used here leverages libraries such as TensorFlow and OpenCV to extract damage features from the images. The extracted data is compared with a historical database on the server. Machine learning algorithms are used for this comparison to determine whether compensation is necessary by calculating similarity.

[0505] Subsequently, the voice and text feedback data sent by the user is analyzed by an emotion recognition module. This analysis utilizes natural language processing (NLP) and voice feature extraction techniques. Specifically, it uses Google's Natural Language API to analyze the tone of emotion.

[0506] If a user's emotions are negative, the server takes this emotional state into consideration and optimizes the notification content. It utilizes a generative AI model to generate information in easy-to-understand language and provides a careful explanation to the user. This process enables not only accurate assessment of damage but also the provision of user-friendly services.

[0507] As a concrete example, an example of a prompt message is shown below.

[0508] "Use an AI model to recognize emotions and suggest how to adjust notifications if the user's feedback indicates dissatisfaction."

[0509] This invention makes it possible to provide users with higher quality and more emotionally sensitive services, contributing to an improved customer experience.

[0510] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0511] Step 1:

[0512] The terminal places the returned equipment in a shooting booth and uses its camera to capture images from multiple angles. The input is the equipment itself, and the output is a JPEG image file. This allows for detailed capture of the equipment's appearance.

[0513] Step 2:

[0514] The device transmits captured images to the server via the network. The input is a JPEG image file, and the output is the image data received by the server. This process is carried out using a communication protocol.

[0515] Step 3:

[0516] The server passes the received image to the AI ​​image analysis module. The input is JPEG image data, and the output is damage feature information. In the image analysis, features of the damaged area are extracted using a convolutional neural network (CNN) with TensorFlow or OpenCV.

[0517] Step 4:

[0518] The server compares the extracted damage features with those from a historical database. The input is the damage feature information, and the output is the compensation decision. A machine learning algorithm calculates the similarity of the feature vectors to determine whether compensation is appropriate.

[0519] Step 5:

[0520] Emotion recognition is performed based on user feedback. The input is voice or text data from the user, and the output is the result of emotion analysis. The server uses natural language processing technology to analyze the emotional tone of the text and perform emotion recognition.

[0521] Step 6:

[0522] The server adjusts the notification content based on the emotion recognition results. The input is the compensation decision result and the emotional state, and the output is a customized notification message. A generative AI model is used to generate explanations that are convincing to the user. This enables information delivery that is considerate of the user.

[0523] (Application Example 2)

[0524] Next, we will explain application example 2. In the following explanation, 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."

[0525] In recent years, consumers expect not only a simple assessment of item damage during returns and exchanges, but also consideration of their feelings and a more satisfactory response. However, current systems struggle to adequately recognize user emotions and respond accordingly at each step of the return process. To improve this situation, there is a need to develop a system that effectively assesses item damage while taking user emotions into account.

[0526] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0527] In this invention, the server includes means for acquiring visual information of returned items, means for analyzing the acquired visual information and extracting characteristics of damage to the items, and means for recognizing the user's emotional state and optimizing the notification of the judgment results. This enables damage assessment and notification that takes the user's emotions into consideration.

[0528] "Returned items" refer to goods that have been sent back by consumers, either used or unused.

[0529] "Visual information" refers to image data acquired using cameras, scanners, etc., and related information based on that data.

[0530] "Analysis" refers to the process of identifying and understanding specific features and patterns based on acquired data.

[0531] "Damage characteristics" refer to the characteristics and morphological changes related to deterioration or damage that have occurred to an item.

[0532] "Damage information" refers to a collection of data on damage to items that have been registered in the past.

[0533] "Eligible for compensation" refers to the criteria for which compensation measures such as repair, replacement, or refund are available if damage or defects are found.

[0534] "Emotional state" refers to the results of measuring and evaluating the psychological and emotional responses exhibited by the user.

[0535] "Optimizing notifications" refers to the process of adjusting the content and method of notifications according to the user's needs and circumstances, and delivering information in the most effective and user-satisfying way.

[0536] The system that realizes this application integrates damage assessment of returned items with user sentiment recognition.

[0537] The server first acquires visual information about the returned item. To do this, it uses the terminal's camera to take images of the item from multiple angles. The captured image data is then transmitted to the server via the internet.

[0538] The server then uses an image analysis module to analyze the visual information and extract features of damage to the object. These image analysis modules include artificial intelligence models such as TensorFlow and OpenCV.

[0539] The server then compares the extracted damage characteristics with previously stored damage information. This comparison determines whether the item is eligible for compensation.

[0540] Furthermore, the server recognizes the user's emotional state from their voice or text. At this stage, emotion analysis is performed using tools such as IBM Watson NLU. Once the emotional state is recognized, the server optimizes the notification content of the judgment result. This results in notifications that include careful explanations that take the user's emotions into consideration.

[0541] For example, if a user expresses feelings of being "very shocked" after returning a product, they will be notified of the appropriate compensation procedures along with a detailed assessment of the damage.

[0542] An example of a prompt for the generating AI model is: "Based on the damage assessment results observed on the camera lens, please advise on how to structure the notification content, taking into account the degree of surprise or dissatisfaction the user is experiencing."

[0543] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0544] Step 1:

[0545] The device takes images of the returned item. It captures visual information by photographing the item from multiple angles. The input is the item itself, and the output is the captured image file. The captured image is then sent to the next step.

[0546] Step 2:

[0547] The server receives the captured image files and sends them to the AI ​​image analysis module. Here, the input is the image data acquired in step 1, and the output is data with the damage features extracted. TensorFlow or OpenCV is used to analyze the damage in the image and capture its features.

[0548] Step 3:

[0549] The server compares the extracted damage characteristics with past damage information in the database. The input is the damage characteristic data obtained in step 2, and the output is the result of the compensation eligibility determination. Compensation is determined by comparing it with similar cases in the past database.

[0550] Step 4:

[0551] The server recognizes the user's emotional state through voice and text feedback. Here, the input is the voice or text feedback provided by the user, and the output is the analyzed emotional state. IBM Watson NLU and similar systems are used to classify the user's emotions.

[0552] Step 5:

[0553] The server optimizes the notification content based on the damage compensation decision and the user's emotional state. The input is the results of steps 3 and 4, and the output is a customized notification message. The notification message is crafted to be sensitive to the user's emotions and is sent in the next step.

[0554] Step 6:

[0555] The server sends the optimized notification content to the user. Here, the input is the notification text generated in step 5, and the output is the notification the user receives. The notification is sent via email or a dedicated application, allowing the user to check it.

[0556] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0557] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0558] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0559] [Fourth Embodiment]

[0560] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0561] As shown in Figure 7, the 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.

[0562] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0563] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0564] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0565] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0566] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0567] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0568] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0569] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0570] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0571] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0572] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0573] The image analysis system of the present invention is built to automate the process of efficiently evaluating returned equipment and analyzing its damage. Specific embodiments of this system are described below.

[0574] Upon arrival at the office, the returned device is placed in a dedicated shooting booth. This booth is equipped with a high-performance camera that captures images of the device from multiple angles. These images are immediately transferred to a server, and the subsequent analysis process begins.

[0575] The server receives the transmitted images and performs analysis within its AI image analysis module. This module utilizes deep learning technology to automatically extract damage features such as cracks and surface scratches from the images. This sophisticated process allows for accurate detection of even minute damage.

[0576] Next, the server compares the feature extraction results with the database. The database contains a collection of past damage data and corresponding reports, and uses this to calculate the similarity of the damage. If the similarity exceeds a certain threshold, the server determines that the damage is eligible for compensation.

[0577] Finally, the server notifies the user of the decision. This notification is sent via email or a dedicated application, and a detailed analysis report is attached to the result. Based on this report, the user can then initiate the compensation process.

[0578] As a concrete example, consider a case where a device is returned and a large crack is found on its screen. The server compares the shape of the crack with its similarity to past data and determines that it is human-caused damage. It then automatically calculates the compensation amount and notifies the user. This entire process ensures operational efficiency and fair judgment.

[0579] The following describes the processing flow.

[0580] Step 1:

[0581] The terminal is placed inside the shooting booth, and a high-resolution camera captures images from multiple angles of the exterior. The captured images are then transmitted directly to the server in digital format.

[0582] Step 2:

[0583] The server inputs the received image into the AI ​​image analysis module and starts the image analysis process. In this process, the server uses deep learning technology to extract damage features such as scratches and cracks.

[0584] Step 3:

[0585] The server compares the extracted features with historical damage data stored in its internal database. Here, it calculates the similarity between the current features and similar damage patterns reported in the past.

[0586] Step 4:

[0587] The server determines whether the damage is covered by compensation based on the similarity calculation results. This determination applies pre-defined criteria to determine whether compensation is necessary depending on the type and extent of the damage.

[0588] Step 5:

[0589] The server notifies the user of the decision. The notification is sent via email or a dedicated application, and a detailed analysis report is attached. This allows the user to immediately initiate the necessary compensation procedures.

[0590] (Example 1)

[0591] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0592] In assessing damage to returned equipment, conventional methods often rely heavily on human judgment, limiting the consistency and speed of evaluations. Furthermore, there is a need for highly accurate analysis of diverse damage patterns. Additionally, there is a lack of effective means to communicate damage assessment results to users, highlighting the need for increased efficiency and fairness in decision-making.

[0593] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0594] In this invention, the server includes means for installing returned equipment and acquiring images from multiple angles in a controlled environment; means for automatically extracting damage features of the equipment using a deep learning algorithm; and means for comparing the acquired damage features with past data and evaluating the similarity. This enables accurate and rapid damage assessment and fair compensation decisions while minimizing human intervention.

[0595] "Returned equipment" refers to devices that have been used by a user and then returned to their original location.

[0596] A "controlled shooting environment" refers to a shooting booth configuration where the light source and shooting conditions are standardized, minimizing the influence of the external environment.

[0597] A "deep learning algorithm" refers to a technology that automatically learns specific features from images through training with large amounts of data, and performs damage detection and classification.

[0598] "Damage characteristics" refer to information about the location, shape, and size of cracks and damage extracted through image analysis.

[0599] "Similarity assessment" refers to the process of comparing extracted damage features with data in past databases and quantifying their correlation.

[0600] A "communication network" refers to a network infrastructure built for the purpose of exchanging data.

[0601] "User" refers to the final recipient of a system or service.

[0602] This invention provides a system for automatically evaluating the damage to returned equipment and efficiently determining whether compensation is applicable. A specific embodiment of this system is described below.

[0603] First, when a user returns a device to the office, it is placed in a dedicated shooting booth. The shooting booth provides a controlled environment, maintaining a constant position for the light source and subject, making it possible to acquire high-definition images every time, eliminating the influence of ambient light. This shooting booth is equipped with a high-performance digital camera that captures detailed images from multiple angles of the device.

[0604] Next, the server quickly receives the captured images and stores them in a central storage device. This data is immediately sent to an AI analysis module, where it is automatically processed using a deep learning algorithm. This algorithm, built using TensorFlow, automatically extracts damage features and recognizes crack and damage patterns with high accuracy. This allows users to quickly detect even minute damage that might be overlooked during human visual inspection.

[0605] Subsequently, the server compares the extracted damage characteristics with past damage information stored in a database. This database records past damage cases, their causes, and repair history, and the server evaluates the similarity to this information to determine compensation. This evaluation result is standardized through an automated process, enabling fair and consistent decisions.

[0606] Ultimately, the server determines whether compensation is applicable and notifies the user of the result. This notification is sent via email or a dedicated application, allowing the user to check the result immediately. The notification includes an analysis report detailing the damage, which the user can use to proceed with the subsequent procedures.

[0607] For example, if a user returns a device with a cracked screen, the server analyzes the shape of the crack by comparing it to past data and determines that it is human-caused damage. At this point, the server automatically calculates the compensation amount and notifies the user. This process allows users to receive compensation quickly and fairly.

[0608] An example of a prompt when using a generative AI model is: "Please analyze the damage to the returned device in detail and inform us about the possibility of compensation."

[0609] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0610] Step 1:

[0611] Upon returning the device, the user places it in a dedicated shooting booth. This booth eliminates ambient light and uses a high-performance camera to capture images of the device from multiple angles. The input data consists of the captured high-resolution images of the device. This results in clear image data being obtained as output.

[0612] Step 2:

[0613] The server receives captured images via the network and stores them in a central repository. The input consists of multiple image files received by the server. The server organizes these image data appropriately and prepares them for immediate use in the subsequent analysis process. The output of this step is the image data ready for analysis.

[0614] Step 3:

[0615] The server passes the saved image to the AI ​​image analysis module and starts the analysis. The input is image data ready for analysis. The server uses a deep learning algorithm powered by TensorFlow to automatically extract damage features such as cracks and scratches from the image. The output includes the extracted damage features as data.

[0616] Step 4:

[0617] The server compares the extracted damage features with a database of past damages and calculates similarity. The inputs are the extracted damage features and the database of past damages. The server uses statistical methods to evaluate the similarity and establish criteria for determining whether compensation is necessary. The output of this step is the damage similarity score.

[0618] Step 5:

[0619] The server makes a compensation decision based on the similarity score and notifies the user of the result. The input is the damage similarity score. The server generates the result and reports it to the user via email or a dedicated application. The output is a notification of the decision result, including a detailed analysis report.

[0620] (Application Example 1)

[0621] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0622] In the damage assessment process for returned goods at logistics centers, there is a need to reduce the variability of human judgment and automate the process for quick and fair compensation decisions. Furthermore, building a system that improves work efficiency by providing real-time notification of results is a key challenge.

[0623] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0624] In this invention, the server includes means for acquiring images of returned items, means for analyzing the acquired images and extracting characteristics of the damage to the items, means for comparing the extracted characteristics with previously registered damage data, means for determining whether the items are eligible for compensation based on the results of the comparison, means for notifying the determination result, and means for the notification means to provide the result in real time via a mobile terminal device. This makes it possible to automate the damage assessment of returned goods at a logistics center and to quickly make fair compensation decisions.

[0625] "Returned items" refers to all products and goods that have been sent back after being used by a customer or user.

[0626] "Means of acquiring images" refers to devices and technologies for visually capturing the whole or parts of an object and saving them as digital data.

[0627] "Means for analyzing images and extracting the characteristics of damage to an item" refers to technologies and devices that process acquired image data to identify damage to the surface or interior of an item and automatically recognize its detailed characteristics.

[0628] "Means of comparing with previously registered damage data" refers to technologies and processes for matching information on previously recorded damage with newly identified damage characteristics.

[0629] "Means of determining whether an item is eligible for compensation" refers to criteria or devices used to determine whether an item meets the compensation requirements, based on the characteristics of the damage and past data.

[0630] "Means of notifying the decision result" refers to communication technologies and systems used to convey information to relevant parties regarding whether compensation is possible and details of the damage.

[0631] "Means of providing results in real time via mobile terminal devices" refers to technologies and systems that allow users to receive decision results immediately through portable digital devices.

[0632] The system implementing this invention incorporates advanced image analysis technology to efficiently assess the damage to returned items.

[0633] First, the items returned by the user are placed in a dedicated shooting environment, and multiple images are acquired from the entire object and from the necessary angles using the camera of a mobile device. This mobile device has a dedicated application installed for transferring the images to a server in the cloud.

[0634] The server performs AI image analysis using the received image data. Specifically, a model utilizing deep learning technology automatically extracts damage features and compares this information with past damage data stored in a database. The AI ​​model employs machine learning techniques using TensorFlow.

[0635] The server determines the need for compensation based on the matching results. If compensation is deemed necessary, the server notifies the user via the mobile device. The notification includes details of the damage and information regarding the application of compensation. The results are provided in real time using the push notification technology used.

[0636] As a concrete example, consider a case where a household electrical appliance returned to a logistics center has a cracked screen due to a fall. This system identifies the damage in real time and immediately determines whether it is eligible for compensation by comparing it with past data.

[0637] An example of a prompt message might be: "From this image, detect the extent of damage to the product upon return and report it by comparing it to similar data." This inquiry is then posed to the generating AI model.

[0638] This will not only dramatically improve operational efficiency at logistics centers, but also ensure fairness and speed in compensation.

[0639] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0640] Step 1:

[0641] The user photographs the returned item using the camera on their mobile device. The user photographs the entire item and any suspected damaged areas from multiple angles and inputs the image data through a dedicated application on the device. The input image data is then transferred to a server in the cloud.

[0642] Step 2:

[0643] The server passes the received image data to the AI ​​analysis module. The server uses a generative AI model based on TensorFlow to analyze the damage features of the object from the image. Here, information extracted as damage features is output in response to the image data input. Specifically, it performs data calculations to detect crack and deformation patterns.

[0644] Step 3:

[0645] The server compares the extracted damage features with historical damage data stored in the database. The server receives the damage information from the analysis as input and calculates its similarity to past compensation cases. This results in an output determining whether the damage is eligible for compensation. Specifically, it performs data matching using a similarity calculation algorithm.

[0646] Step 4:

[0647] The server notifies the mobile device of the assessment result and communicates it to the user. The notification from the server includes detailed information about the damage and the assessment result regarding compensation, and is pushed to the user in real time. The output of the notification serves as a basis for the user to decide whether to file a compensation claim. Specifically, the result is sent via a notification API.

[0648] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0649] This invention provides a system that combines user emotion recognition with a device damage assessment process. This system not only analyzes the damage to returned devices using image analysis and makes compensation decisions based on the analysis results, but also recognizes the user's emotional state and optimizes the notification process.

[0650] First, the returned device is placed in a shooting booth, and images are automatically taken from multiple angles. The captured images are immediately sent to a server. This image data is analyzed by an AI image analysis module, and damage characteristics are extracted. The server compares the extracted characteristics with an internal historical database to determine whether the damage is eligible for compensation.

[0651] Subsequently, the server considers the user's emotions when notifying them of the decision. This function is provided by an emotion recognition module, which analyzes the user's voice and written feedback data. If the user's emotions are negative, such as dissatisfaction or surprise, the server adjusts the notification content to include a polite and convincing explanation before sending it.

[0652] For example, if a user is dissatisfied with device damage, emotion recognition can detect that emotion. In this case, the server sends a report in addition to the usual notification, which includes a detailed explanation of the compensation and information on how the damage was assessed. This demonstrates consideration for the user and improves the customer experience.

[0653] This invention makes it possible to provide higher quality service not only by accurately assessing damage, but also by incorporating user emotional responses.

[0654] The following describes the processing flow.

[0655] Step 1:

[0656] Once the device is returned, it is placed in the shooting booth. A high-resolution camera captures images of various parts of the device from multiple angles, and the acquired images are sent to a server as digital data.

[0657] Step 2:

[0658] The server inputs the received image data into an AI image analysis module. This module extracts damage features such as scratches and cracks from the image. The extracted features are detailed data indicating the type, location, and extent of the damage.

[0659] Step 3:

[0660] The server compares the extracted damage characteristics with a historical database and calculates the similarity. The database contains previously recorded damage cases, and the server uses this comparison to determine whether the damage is eligible for compensation.

[0661] Step 4:

[0662] Based on the decision, the server creates notification content while considering the user's emotional state. The emotion recognition module analyzes the voice data and feedback obtained from the user to determine the user's emotions.

[0663] Step 5:

[0664] The server responds flexibly to the user's emotional state. For example, if negative emotions are detected, the notification will include additional information about compensation details and the decision-making process, demonstrating consideration for the user.

[0665] Step 6:

[0666] Finally, the server sends the assessment results and adjusted notification to the user. This allows the user to accurately understand the damage assessment results and compensation details and proceed with the necessary procedures.

[0667] (Example 2)

[0668] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0669] Conventional damage assessment systems rely solely on objective data regarding the condition of returned equipment when determining damage, which means they cannot take into account user emotions or feedback. This results in a lack of consideration for the user in notifying them of the results, making it difficult to provide an optimal customer experience.

[0670] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0671] In this invention, the server includes means for acquiring multi-angle images of returned equipment, means for transmitting the acquired image data to a processing device, means for extracting damage features from the transmitted image data using analysis techniques, means for comparing the extracted features with past data to evaluate the damage, means for determining whether compensation is applicable based on the evaluation results, and means for optimizing the content of the transmission when communicating the judgment results, taking into account the user's emotional state. This enables accurate damage evaluation, allows for notifications that take the user's emotions into consideration, and provides an improved customer experience.

[0672] "Returned equipment" refers to electronic devices and equipment that have been returned after being used by a user.

[0673] "Multi-angle images" refer to multiple images of an object taken from different angles, providing three-dimensional information.

[0674] A "processing device" refers to a computer or server device that receives data and performs analysis and evaluation.

[0675] "Analysis techniques" refer to algorithms and methods for extracting meaningful information from data, and include image analysis and natural language processing.

[0676] "Damage characteristics" refer to specific data points or patterns that indicate physical defects or abnormalities that have occurred in the equipment.

[0677] "Past data" refers to stored information in a database that includes information about damage collected previously.

[0678] "Damage assessment" refers to the process of determining the degree and importance of equipment damage based on extracted characteristics.

[0679] "Eligible for compensation" refers to equipment that has been determined to be covered by warranty or repair, depending on the extent and type of damage.

[0680] "User's emotional state" refers to the emotional aspects of the feedback provided by the device user, including states such as dissatisfaction and satisfaction.

[0681] "Optimizing message content" refers to the process of adjusting the content of the message being conveyed, taking into account the emotional state of the user.

[0682] This invention provides a system that delivers more accurate and user-friendly notifications by combining damage assessment of returned devices with user emotion recognition. Specific embodiments are described below.

[0683] The terminal places the returned device in a dedicated shooting booth. This shooting booth is equipped with cameras that can photograph the device from multiple angles, thereby automatically acquiring multiple images. The images are saved in JPEG format and sent to a server via the network.

[0684] The server supplies the received image data to an AI image analysis module. The AI ​​technology used here leverages libraries such as TensorFlow and OpenCV to extract damage features from the images. The extracted data is compared with a historical database on the server. Machine learning algorithms are used for this comparison to determine whether compensation is necessary by calculating similarity.

[0685] Subsequently, the voice and text feedback data sent by the user is analyzed by an emotion recognition module. This analysis utilizes natural language processing (NLP) and voice feature extraction techniques. Specifically, it uses Google's Natural Language API to analyze the tone of emotion.

[0686] If a user's emotions are negative, the server takes this emotional state into consideration and optimizes the notification content. It utilizes a generative AI model to generate information in easy-to-understand language and provides a careful explanation to the user. This process enables not only accurate assessment of damage but also the provision of user-friendly services.

[0687] As a concrete example, an example of a prompt message is shown below.

[0688] "Use an AI model to recognize emotions and suggest how to adjust notifications if the user's feedback indicates dissatisfaction."

[0689] This invention makes it possible to provide users with higher quality and more emotionally sensitive services, contributing to an improved customer experience.

[0690] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0691] Step 1:

[0692] The terminal places the returned equipment in a shooting booth and uses its camera to capture images from multiple angles. The input is the equipment itself, and the output is a JPEG image file. This allows for detailed capture of the equipment's appearance.

[0693] Step 2:

[0694] The device transmits captured images to the server via the network. The input is a JPEG image file, and the output is the image data received by the server. This process is carried out using a communication protocol.

[0695] Step 3:

[0696] The server passes the received image to the AI ​​image analysis module. The input is JPEG image data, and the output is damage feature information. In the image analysis, features of the damaged area are extracted using a convolutional neural network (CNN) with TensorFlow or OpenCV.

[0697] Step 4:

[0698] The server compares the extracted damage features with those from a historical database. The input is the damage feature information, and the output is the compensation decision. A machine learning algorithm calculates the similarity of the feature vectors to determine whether compensation is appropriate.

[0699] Step 5:

[0700] Emotion recognition is performed based on user feedback. The input is voice or text data from the user, and the output is the result of emotion analysis. The server uses natural language processing technology to analyze the emotional tone of the text and perform emotion recognition.

[0701] Step 6:

[0702] The server adjusts the notification content based on the emotion recognition results. The input is the compensation decision result and the emotional state, and the output is a customized notification message. A generative AI model is used to generate explanations that are convincing to the user. This enables information delivery that is considerate of the user.

[0703] (Application Example 2)

[0704] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0705] In recent years, consumers expect not only a simple assessment of item damage during returns and exchanges, but also consideration of their feelings and a more satisfactory response. However, current systems struggle to adequately recognize user emotions and respond accordingly at each step of the return process. To improve this situation, there is a need to develop a system that effectively assesses item damage while taking user emotions into account.

[0706] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0707] In this invention, the server includes means for acquiring visual information of returned items, means for analyzing the acquired visual information and extracting characteristics of damage to the items, and means for recognizing the user's emotional state and optimizing the notification of the judgment results. This enables damage assessment and notification that takes the user's emotions into consideration.

[0708] "Returned items" refer to goods that have been sent back by consumers, either used or unused.

[0709] "Visual information" refers to image data acquired using cameras, scanners, etc., and related information based on that data.

[0710] "Analysis" refers to the process of identifying and understanding specific features and patterns based on acquired data.

[0711] "Damage characteristics" refer to the characteristics and morphological changes related to deterioration or damage that have occurred to an item.

[0712] "Damage information" refers to a collection of data on damage to items that have been registered in the past.

[0713] "Eligible for compensation" refers to the criteria for which compensation measures such as repair, replacement, or refund are available if damage or defects are found.

[0714] "Emotional state" refers to the results of measuring and evaluating the psychological and emotional responses exhibited by the user.

[0715] "Optimizing notifications" refers to the process of adjusting the content and method of notifications according to the user's needs and circumstances, and delivering information in the most effective and user-satisfying way.

[0716] The system that realizes this application integrates damage assessment of returned items with user sentiment recognition.

[0717] The server first acquires visual information about the returned item. To do this, it uses the terminal's camera to take images of the item from multiple angles. The captured image data is then transmitted to the server via the internet.

[0718] The server then uses an image analysis module to analyze the visual information and extract features of damage to the object. These image analysis modules include artificial intelligence models such as TensorFlow and OpenCV.

[0719] The server then compares the extracted damage characteristics with previously stored damage information. This comparison determines whether the item is eligible for compensation.

[0720] Furthermore, the server recognizes the user's emotional state from their voice or text. At this stage, emotion analysis is performed using tools such as IBM Watson NLU. Once the emotional state is recognized, the server optimizes the notification content of the judgment result. This results in notifications that include careful explanations that take the user's emotions into consideration.

[0721] For example, if a user expresses feelings of being "very shocked" after returning a product, they will be notified of the appropriate compensation procedures along with a detailed assessment of the damage.

[0722] An example of a prompt for the generating AI model is: "Based on the damage assessment results observed on the camera lens, please advise on how to structure the notification content, taking into account the degree of surprise or dissatisfaction the user is experiencing."

[0723] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0724] Step 1:

[0725] The device takes images of the returned item. It captures visual information by photographing the item from multiple angles. The input is the item itself, and the output is the captured image file. The captured image is then sent to the next step.

[0726] Step 2:

[0727] The server receives the captured image files and sends them to the AI ​​image analysis module. Here, the input is the image data acquired in step 1, and the output is data with the damage features extracted. TensorFlow or OpenCV is used to analyze the damage in the image and capture its features.

[0728] Step 3:

[0729] The server compares the extracted damage characteristics with past damage information in the database. The input is the damage characteristic data obtained in step 2, and the output is the result of the compensation eligibility determination. Compensation is determined by comparing it with similar cases in the past database.

[0730] Step 4:

[0731] The server recognizes the user's emotional state through voice and text feedback. Here, the input is the voice or text feedback provided by the user, and the output is the analyzed emotional state. IBM Watson NLU and similar systems are used to classify the user's emotions.

[0732] Step 5:

[0733] The server optimizes the notification content based on the damage compensation decision and the user's emotional state. The input is the results of steps 3 and 4, and the output is a customized notification message. The notification message is crafted to be sensitive to the user's emotions and is sent in the next step.

[0734] Step 6:

[0735] The server sends the optimized notification content to the user. Here, the input is the notification text generated in step 5, and the output is the notification the user receives. The notification is sent via email or a dedicated application, allowing the user to check it.

[0736] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0737] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0738] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0739] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0740] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0741] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0742] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0743] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0744] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0745] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0746] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0747] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0748] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0749] 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.

[0750] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0751] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0752] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0753] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0754] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0755] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0756] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0757] The following is further disclosed regarding the embodiments described above.

[0758] (Claim 1)

[0759] Means for obtaining images of the returned child,

[0760] Means for analyzing the acquired images and extracting characteristics of damage to the device,

[0761] A means for comparing the extracted features with previously registered damage data,

[0762] A means of determining whether compensation is applicable based on the results of the above comparison,

[0763] Means of notifying the decision result,

[0764] A system that includes this.

[0765] (Claim 2)

[0766] A system according to claim 1, wherein the means for analyzing the image is to perform feature extraction using a deep learning model.

[0767] (Claim 3)

[0768] The system according to claim 1, wherein the notification means notifies the decision result via email or a dedicated application.

[0769] "Example 1"

[0770] (Claim 1)

[0771] A means of installing the returned equipment and acquiring images from multiple angles in a controlled shooting environment,

[0772] A means for receiving the acquired image and automatically extracting damage features of the equipment using a deep learning algorithm,

[0773] A means for comparing the extracted damage characteristics with previously stored damage information and evaluating the similarity,

[0774] A means for determining whether compensation is applicable based on the aforementioned similarity assessment,

[0775] A means for reporting the aforementioned judgment result via a communication network,

[0776] A system that includes this.

[0777] (Claim 2)

[0778] The system according to claim 1, wherein the controlled shooting environment has a function to maintain constant light conditions.

[0779] (Claim 3)

[0780] The system according to claim 1, wherein the reporting means transmits an analysis report to the user via a network.

[0781] "Application Example 1"

[0782] (Claim 1)

[0783] Means for obtaining images of returned items,

[0784] A means for analyzing the acquired images and extracting the characteristics of damage to the article,

[0785] A means for comparing the extracted features with previously registered damage data,

[0786] A means of determining whether compensation is applicable based on the results of the above comparison,

[0787] Means for notifying the aforementioned judgment result,

[0788] The notification means includes means for providing results in real time via a mobile terminal device,

[0789] A system that includes this.

[0790] (Claim 2)

[0791] The system according to claim 1, wherein the means for analyzing the aforementioned image is to perform feature extraction using machine learning technology.

[0792] (Claim 3)

[0793] The system according to claim 1, wherein the notification means transmits the determination result using a mobile communication terminal or a data processing device.

[0794] "Example 2 of combining an emotion engine"

[0795] (Claim 1)

[0796] A means for acquiring multi-angle images of returned equipment,

[0797] Means for transmitting the acquired image data to a processing device,

[0798] A means for extracting damage characteristics from the transmitted image data using an analysis technique,

[0799] A means for comparing the extracted features with past data and evaluating the damage,

[0800] A means of determining whether compensation is applicable based on the results of the aforementioned evaluation,

[0801] When communicating the results of a decision, a means to optimize the content of the message, taking into account the user's emotional state,

[0802] A system that includes this.

[0803] (Claim 2)

[0804] The system according to claim 1, wherein the means for extracting the characteristics of the damage is to perform analysis using a machine learning model.

[0805] (Claim 3)

[0806] The system according to claim 1, wherein the means for analyzing the user's emotional state performs emotional analysis based on voice or text data.

[0807] "Application example 2 when combining with an emotional engine"

[0808] (Claim 1)

[0809] Means for obtaining visual information of returned items,

[0810] A means for analyzing the acquired visual information and extracting the characteristics of damage to the item,

[0811] A means for comparing the extracted features with previously registered damage information,

[0812] A means of determining whether compensation is applicable based on the results of the above comparison,

[0813] A means for recognizing the user's emotional state and optimizing the notification of the judgment result,

[0814] A system that includes this.

[0815] (Claim 2)

[0816] The system according to claim 1, wherein the means for analyzing the visual information is to perform feature extraction using an artificial intelligence model.

[0817] (Claim 3)

[0818] The system according to claim 1, wherein the notification means notifies the decision result via electronic communication means, and the content of the notification includes an explanation based on the user's emotional state. [Explanation of symbols]

[0819] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means for obtaining images of returned items, A means for analyzing the acquired images and extracting the characteristics of damage to the article, A means for comparing the extracted features with previously registered damage data, A means of determining whether compensation is applicable based on the results of the above comparison, Means for notifying the aforementioned judgment result, The notification means includes means for providing results in real time via a mobile terminal device, A system that includes this.

2. The system according to claim 1, wherein the means for analyzing the aforementioned image is to perform feature extraction using machine learning technology.

3. The system according to claim 1, wherein the notification means transmits the determination result using a mobile communication terminal or a data processing device.

Images