system

Abstract

To provide a system for improving the accuracy of examination by finding an illegal application in an early stage in an online application.SOLUTION: The system includes a means for extracting text data from the identification document using an OCR technique, a means for analyzing features of the extracted text data using an AI algorithm, a means for comparing the analyzed features of the text data with an existing database to determine authenticity, and a means for notifying an examination result based on the determination result.SELECTED DRAWING: Figure 11

Description

[Technical Field]

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

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

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

[0004] The use of forged or altered identification documents in online applications is on the rise, resulting in a decline in the quality of screening and serious financial risks to companies due to fraudulent applications. Conventional systems primarily rely on human visual screening, which can only check a portion of the total, making it difficult to completely prevent fraudulent applications. Therefore, it is necessary to detect fraudulent applications early and improve the accuracy of screening. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides a system that uses the following means. First, it provides a means for extracting character data from identification documents using OCR technology. Next, it provides a means for analyzing the characteristics of the extracted character data using an AI algorithm. Then, it provides a means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity. It also provides a means for notifying the review results based on the results of that determination. It also provides a preprocessing means for resizing the image data, removing noise, correcting tilt, and adjusting brightness and contrast to improve the accuracy of the OCR and AI processing. Finally, it provides a classification means for classifying the results as "true," "false," or "suspicious" based on the analysis results, allowing the system to send the results to a visual review if necessary. This prevents fraudulent applications and improves the accuracy of the review and the safety of the company.

[0006] "OCR technology" is a technology that automatically extracts text information from images and scanned documents.

[0007] An "AI algorithm" is a computational procedure for analyzing large amounts of data and performing pattern recognition and predictions.

[0008] "Identification document" means an official document used to prove an individual's identity.

[0009] "Text data" refers to text information contained in a document that has been converted into a digital format.

[0010] "Preprocessing" refers to the initial processing performed to improve data quality, and includes noise removal and image resizing.

[0011] "Determination result" refers to the result of analysis using AI and OCR technology to distinguish between authenticity and counterfeit.

[0012] "Visual inspection" means visual inspection and evaluation by the human eye.

[0013] A "database" is a system for efficiently storing, managing, and searching data.

[0014] "Feature analysis" is a technique for revealing specific patterns or characteristics from collected data.

[0015] "Resizing" refers to the process of changing the size of data or images. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0038] System configuration and operation overview

[0039] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[0040] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0041] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The result is also notified to the application system, and the results are shared with the user.

[0042] Specific example of system operation

[0043] 1. User Actions

[0044] A user applies for a new mobile phone contract online.

[0045] The user uploads an image of their driver's license and enters the required personal information.

[0046] 2. Server Processing

[0047] The server receives image data from the user and stores it in a database.

[0048] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0049] The preprocessed image is fed to an OCR engine to extract text data.

[0050] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0051] 3. Judgment and Notification

[0052] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0053] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0054] The server notifies the application system of the judgment result and displays the examination result to the user.

[0055] This system automatically verifies the authenticity of identification documents, enabling early detection of fraudulent applications and improving the accuracy of screening. Furthermore, it is possible to reduce financial risks for companies and significantly improve the efficiency of application processing. The system based on this invention can be easily incorporated into online application systems and is applicable to a wide range of uses.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[0059] Step 2:

[0060] The server receives the image data sent by the user and stores the received image data in a database.

[0061] Step 3:

[0062] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[0063] Image resizing: Adjust the size so that OCR and AI can process it properly.

[0064] Denoise: Apply filtering to reduce noise in the image.

[0065] Deskew: If the image is tilted, correct the tilt.

[0066] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[0067] Step 4:

[0068] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. The extracted data includes name, address, date of birth, etc. The server saves the extracted text data as a text file.

[0069] Step 5:

[0070] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[0071] Step 6:

[0072] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[0073] Step 7:

[0074] The server performs the following classification based on the comparison results:

[0075] "True": If there is a high degree of match with existing data.

[0076] "False": If the match is very low.

[0077] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[0078] Step 8:

[0079] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[0080] Step 9:

[0081] The application site will be set up so that users can check the results of the screening. Users can check on the website whether their application has been approved or not.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] Conventional identity verification methods require manual verification of document authenticity, which takes a great deal of time and effort. It also makes it difficult to detect fraudulent applications early, resulting in low accuracy and efficiency of screening. Furthermore, with the spread of online applications, there is a growing need for digitization and automated processing of documents, but current systems are inadequate to meet this demand.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0086] In this invention, the server includes: means for extracting character data from identification documents using OCR technology; means for analyzing the characteristics of the extracted character data using an AI algorithm; means for comparing the analyzed character data characteristics with an existing database to determine authenticity; means for notifying the review results based on the results of the review; means for the user to upload an image of the identification document to the online application system using a terminal; means for the server to store and preprocess the received image data; means for providing the preprocessed image to an OCR engine to extract text data; means for analyzing the character arrangement, font, position, size, etc. using an AI algorithm; means for calculating the degree of match based on the analysis results and classifying the result as "true," "false," or "suspicious"; and means for notifying a manual review team if the result is suspicious. This enables automatic and highly accurate determination of the authenticity of identification documents and early detection of fraudulent applications. It also improves the efficiency and accuracy of review and can be easily incorporated into online application systems.

[0087] "OCR technology" is optical character recognition technology, which extracts character data from images.

[0088] "AI algorithm" refers to a calculation method or model that uses artificial intelligence, and is a technology used to analyze and process data.

[0089] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[0090] "Image data" refers to image information stored in digital format, and includes images of identification documents.

[0091] "Preprocessing" refers to processing performed to analyze image data in optimal conditions, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[0092] "Text data" refers to character information extracted from image data and is obtained using OCR technology.

[0093] "Feature analysis" is the process of analyzing various aspects or characteristics of data, including character alignment, font, position, size, etc.

[0094] "Database" refers to a system or collection for systematically managing and storing large amounts of data, in which authentic identity verification data is stored.

[0095] "Concordance" is an index that indicates how closely the analysis results match the records in the database.

[0096] A "judgment result" is a conclusion based on analysis and degree of agreement, and is classified as "true," "false," or "doubtful."

[0097] The "Manual Review Team" is a specialized team that performs detailed human review of suspicious judgment results.

[0098] "Online application system" refers to a system that allows various applications and procedures to be carried out via the Internet.

[0099] A "server" is a computer system that stores, processes, and provides data over a network.

[0100] "Terminal" refers to a device such as a computer or smartphone operated by a user.

[0101] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0102] System configuration

[0103] The main components of the system are:

[0104] 1. User device (PC or smartphone)

[0105] 2. Server

[0106] 3. Online Application System

[0107] 4. OCR engine (e.g. Tesseract OCR)

[0108] 5. AI Algorithms

[0109] 6. Database

[0110] Operation overview

[0111] A user uses a terminal to upload an image of their identification document, such as a driver's license, through an online application system. The server then stores the received image data. The stored image data is preprocessed to resize, remove noise, straighten, and adjust brightness and contrast. This preprocessing is necessary to improve the accuracy of OCR analysis.

[0112] After preprocessing, the image is analyzed by an OCR engine to extract text data. The server saves this text data as a text file. The extracted text data is then input into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0113] Judgment and Notification

[0114] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, it is notified to the manual review team, who then conduct a visual review. Finally, the result is notified to the application system, and the results are shared with the user.

[0115] Specific operation example

[0116] User operation

[0117] A user applies for a new mobile phone contract online.

[0118] The user uploads an image of their driver's license and enters the required personal information.

[0119] Server processing

[0120] The server receives image data from the user and stores it in a database.

[0121] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0122] The preprocessed image is fed to an OCR engine to extract text data.

[0123] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0124] Judgment and Notification

[0125] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0126] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0127] The server notifies the application system of the judgment result and displays the examination result to the user.

[0128] As a concrete example, the following prompt sentence is input to a generative AI model:

[0129] Example prompt sentence:

[0130] Please explain in detail the processing steps of the system for determining the authenticity of ID documents. The user uploads an image of their driver's license, the server performs preprocessing and OCR analysis, and then an AI algorithm analyzes the features to determine authenticity.

[0131] This system automatically determines the authenticity of identification documents, significantly improving the accuracy and efficiency of screening. It also enables early detection of fraudulent applications, reducing financial risks for companies. The system based on this invention can be easily incorporated into existing online application systems and is applicable to a wide range of uses.

[0132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0133] Step 1:

[0134] User upload of identification documents

[0135] A user uses a terminal (PC or smartphone) to access the online application system and uploads an image of an identification document (e.g., a driver's license).

[0136] Input: Image of user's ID document

[0137] Output: Image data uploaded to the online application system

[0138] Step 2:

[0139] Receiving and storing image data by the server

[0140] The server receives the image data uploaded from the online application system. The received image data is stored on the server, and metadata linked to the user ID and application details is also recorded in the image data.

[0141] Input: Uploaded image data and metadata

[0142] Output: Stored image data and associated metadata

[0143] Step 3:

[0144] Preprocessing of image data by the server

[0145] The server performs pre-processing on the stored image data, which includes the following specific operations:

[0146] Resize: Unify image resolution to improve processing efficiency.

[0147] Noise Reduction: Remove unwanted noise in the image.

[0148] Deskew: Corrects the tilt of the image so that the text is level.

[0149] Brightness and Contrast Adjustment: Properly adjust the brightness and contrast of the image to improve the accuracy of OCR analysis.

[0150] Input: Saved image data

[0151] Output: Preprocessed image data

[0152] Step 4:

[0153] Server-based OCR analysis and text data extraction

[0154] After preprocessing, the image data is input to an OCR engine (e.g., Tesseract OCR) to extract text data from the image. The server saves this text data as a text file.

[0155] Input: Preprocessed image data

[0156] Output: Extracted text data

[0157] Step 5:

[0158] Feature analysis using AI algorithms on the server

[0159] The server inputs the extracted text data into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results provide detailed feature information about the text data.

[0160] Input: Extracted text data

[0161] Output: Analyzed feature information

[0162] Step 6:

[0163] Comparison with database by server and calculation of match

[0164] The server compares the analysis results with an existing database and calculates the degree of match, which determines how closely the text matches the authentic text.

[0165] Input: Analyzed feature information and normal data in the database

[0166] Output: Match rating

[0167] Step 7:

[0168] Classification and notification of judgment results by the server

[0169] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is "suspicious," it sends an alert to the manual review team. Finally, the result is notified to the online application system, which then shares the results with the user.

[0170] Input: Match evaluation result

[0171] Output: Classified judgment results and notification information

[0172] For example, when a user applies for a new mobile phone contract, they upload an image of their driver's license. The server receives the image, performs preprocessing and OCR analysis, and then performs AI feature analysis and comparison with a database. If the AI ​​algorithm determines the image to be "suspicious," an alert is sent to the manual review team. After this process, the user is finally notified of the result.

[0173] (Application example 1)

[0174] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0175] Conventional identity verification systems require manual verification, which requires a great deal of time and resources. They also lack reliable methods for preventing the use of fraudulent identity documents. While electronic payment services require the immediate registration of new accounts, they also require highly accurate verification of the authenticity of identity documents. Therefore, there is a growing need for a system that can efficiently and automatically verify the authenticity of identity documents and immediately notify users of the results.

[0176] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0177] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for determining authenticity by comparing the characteristics of the analyzed character data with an existing database, means for notifying the result of the examination based on the determination result, means for transmitting image data of the identification document from the mobile device to the server at the time of new registration, and means for extracting text data from the preprocessed image data and analyzing the characteristics of the character string using a model. This makes it possible to efficiently and automatically determine the authenticity of identification documents in electronic payment services and immediately notify the user of the result.

[0178] "OCR technology" stands for optical character recognition technology, which is a technology that automatically extracts characters from images.

[0179] An "AI algorithm" is an algorithm that uses artificial intelligence, and is a method of automatically learning and making decisions based on large amounts of data.

[0180] "Identification document" refers to an official document used to verify an individual's identity, such as a driver's license or passport.

[0181] "Text data" refers to text information extracted from an image using OCR technology.

[0182] "Feature analysis" is the process of analyzing the characteristics of extracted character data, such as the arrangement, font, position, and size.

[0183] The "existing database" is a database that stores previously accumulated characteristic data of identification documents.

[0184] "Authenticity determination" is the process of determining whether a document is genuine or fake by comparing the characteristics of the analyzed character data with existing databases.

[0185] "Notification of inspection results" is the process of notifying the user or system of the results of the authenticity determination.

[0186] A "mobile device" is a portable electronic device such as a smartphone or tablet.

[0187] A "server" is a computer system that processes data, analyzes image and text data, and manages the results of judgments.

[0188] "Preprocessing" is the process of resizing, removing noise, correcting tilt, and adjusting brightness and contrast to make image data easier to analyze.

[0189] "Model" means a machine learning model trained by an AI algorithm and used to analyze data.

[0190] This invention is a system that efficiently and automatically determines the authenticity of identity documents in electronic payment services and immediately notifies the user of the results. This system eliminates the need for the conventional manual verification process and enables rapid and accurate identity verification.

[0191] System configuration and operation overview

[0192] 1. User Action:

[0193] When a user creates a new account for an electronic payment service, the user takes a picture of their identification document (e.g., driver's license) using a mobile device such as a smartphone and sends it to a server via an application.

[0194] 2. Server Processing:

[0195] The server receives the image data of the ID document sent by the user and performs preprocessing, including image resizing, noise removal, tilt correction, brightness and contrast adjustment, etc. This is done using OpenCV, an open source image processing library.

[0196] After preprocessing, the image is then processed using the OCR engine Pytesseract to extract text data, which contains the necessary character information and is then analyzed using AI algorithms.

[0197] The extracted character data is fed into a pre-trained machine learning model, which is trained using a machine learning library such as scikit-learn, to analyze specific features of the string (e.g., character alignment, font, position, size, etc.).

[0198] 3. Authentication and Notification:

[0199] The server compares the characteristics of the analyzed text data with an existing database and calculates the degree of match. Based on this degree of match, the authenticity is classified into three categories: "true," "false," or "suspicious." If the authenticity is doubtful, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified.

[0200] The result of the check is immediately notified to the user via the application, for example, a message such as "Your identity verification documents are valid, so your account creation has been completed."

[0201] Specific use cases

[0202] Register a new account:

[0203] A user launches an application to create a new electronic payment account and uploads a photo of their driver's license.

[0204] The server receives the images, pre-processes them, and analyzes them using OCR and AI algorithms.

[0205] The result of the judgment is immediately notified to the user, and if it is true, the account creation is completed.

[0206] Prompt Sentence Examples

[0207] To create a new e-payment account, please upload a photo of your driver's license. The system will automatically verify the authenticity of your ID and notify you of the processing result.

[0208] The system is expected to speed up the process of registering new accounts for electronic payment services, improve the accuracy of identity verification, and significantly improve the user experience.

[0209] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0210] Step 1:

[0211] To create a new electronic payment account, a user takes a photo of their identity document with their mobile device and uploads the image to a server via an application. The input is the image data of the identity document, which is then sent to the server. The output is the image data stored on the server.

[0212] Step 2:

[0213] The server performs preprocessing on the received image data. Specifically, it uses OpenCV to resize, remove noise, correct distortion, and adjust brightness and contrast. The input is the uploaded image data, and the output is the preprocessed image data.

[0214] Step 3:

[0215] The server supplies the preprocessed image data to an OCR engine (Pytesseract) to extract character data. The input is the preprocessed image data, and the output is the extracted character data. At this stage, character information is obtained in text format.

[0216] Step 4:

[0217] The extracted character data is input into an AI algorithm for feature analysis. Specifically, a machine learning library (such as scikit-learn) is used to analyze features such as character arrangement, font, position, and size. The input is text data, and the output is the analyzed feature data.

[0218] Step 5:

[0219] The server compares the analyzed data with the existing database and calculates the degree of match. The input is the analyzed data and the existing database, and the output is a score of the degree of match. In this step, an authentication algorithm based on a specific model is used.

[0220] Step 6:

[0221] Based on the degree of match, the authenticity is judged and the judgment result is classified into three categories: "true," "false," or "suspicious." The input is the match score, and the output is the status of the judgment result. In this step, if the result is suspicious, a flag is set indicating that manual review is required.

[0222] Step 7:

[0223] The server notifies the user of the final review result through the application system. Specifically, the review result is displayed on the user's mobile device. The input is the review result status, and the output is the review result displayed to the user. This result allows the user to confirm whether or not the account creation is complete.

[0224] The above steps effectively automate the new account registration process for electronic payment services.

[0225] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0226] This invention combines an emotion engine with a system for automatically determining the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It then combines an emotion engine that recognizes the user's emotional state and optimizes the notification process of the screening results based on the emotional information.

[0227] System configuration and operation overview

[0228] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[0229] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0230] Based on this degree of match, the server classifies the judgment result as "true," "false," or "suspicious." If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The judgment result is also notified to the application system, and the result is shared with the user. At the same time, the emotion engine analyzes the user's emotional state and reflects this information in the inspection process.

[0231] Overview of Emotion Engine Operation

[0232] The emotion engine analyzes the user's emotional state in real time based on camera images, audio data, timing information of input operations, etc. The results of this analysis are sent to the server and used to notify the review results and for follow-up procedures.

[0233] Specific example of system operation

[0234] 1. User Actions

[0235] A user applies for a new mobile phone contract online.

[0236] The user uploads an image of their driver's license and enters the required personal information.

[0237] 2. Server Processing

[0238] The server receives image data from the user and stores it in a database.

[0239] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0240] The preprocessed image is fed to an OCR engine to extract text data.

[0241] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0242] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0243] 3. Coordination of judgment and emotion analysis

[0244] The server classifies the results as "true," "false," or "suspicious" based on the degree of match.

[0245] The server starts the emotion engine and collects and analyzes the user's emotion data.

[0246] The server integrates the analysis results and sends the review results to the application system. Emotional data is used to customize notification methods and message content.

[0247] 4. Notice to Users

[0248] The user checks the results of the review on the application site, and the notification includes appropriate feedback that reflects the results of the sentiment analysis.

[0249] If necessary, follow-up suggestions are provided to the user.

[0250] This system based on the present invention not only improves the accuracy of determining the authenticity of identification documents, but also improves the user experience and further prevents fraud by enabling flexible notification of review results that take into account the user's emotional state.

[0251] The processing flow will be explained below.

[0252] Step 1:

[0253] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[0254] Step 2:

[0255] The server receives the image data sent by the user and stores the received image data in a database.

[0256] Step 3:

[0257] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[0258] Image resizing: Adjust the size so that OCR and AI can process it properly.

[0259] Denoise: Apply filtering to reduce noise in the image.

[0260] Deskew: If the image is tilted, correct the tilt.

[0261] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[0262] Step 4:

[0263] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. This data includes the name, address, date of birth, etc. The server saves the extracted text data as a text file for further processing.

[0264] Step 5:

[0265] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[0266] Step 6:

[0267] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[0268] Step 7:

[0269] The server performs the following classification based on the comparison results:

[0270] "True": If there is a high degree of match with existing data.

[0271] "False": If the match is very low.

[0272] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[0273] Step 8:

[0274] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[0275] Step 9:

[0276] The server starts the emotion engine and collects camera images, audio data, and timing information of input operations from the user's device. The emotion engine uses this data to analyze the user's emotional state in real time.

[0277] Step 10:

[0278] The server integrates the emotion analysis results into the review result notification process, and optimizes the content and method of review result notification taking into account the user's emotional state.

[0279] Step 11:

[0280] The server sends the final review results, feedback based on the emotional state, and follow-up suggestions, if necessary, to the application system.

[0281] Step 12:

[0282] The user checks the results of the review on the application site. The notification includes appropriate feedback that reflects the results of the sentiment analysis. The user can then check the results and take the next action.

[0283] Example 2

[0284] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0285] In conventional identity verification systems, determining the authenticity of identity documents is often done manually, which is time-consuming and labor-intensive. Even in automated systems, there is no notification process that takes into account the user's emotional state, making it difficult to improve the user experience. Furthermore, the lack of emotion analysis means that notification of results is uniform, making it difficult to respond flexibly.

[0286] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0287] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity, emotion engine means for analyzing the user's emotional state in real time, means for optimizing the notification process of the examination result based on the analysis result of the emotion engine means, and means for notifying the examination result based on the determination result. This improves the accuracy of authenticity determination and enables a flexible notification process based on the user's emotional state, significantly improving the user experience.

[0288] "OCR technology" stands for optical character recognition technology, which recognizes characters from image data and extracts them as text data.

[0289] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[0290] An "AI algorithm" is an algorithm based on artificial intelligence technology that learns, analyzes, and predicts specific tasks.

[0291] An "emotion engine" is a system for analyzing a user's emotional state, and is a technology that recognizes emotions in real time based on input information such as camera images and audio data.

[0292] "Preprocessing" refers to processing performed to prepare image data in a format that is easier to analyze, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[0293] "Resizing" is an operation for changing the resolution and size of image data, and is a process for adjusting the resolution to a specified value.

[0294] "Noise reduction" is a process that removes unnecessary information and dust from image data, and is performed to improve image quality.

[0295] "Tilt correction" is a process of correcting the tilt of image data and leveling it.

[0296] "Brightness and contrast adjustment" is a process for adjusting the brightness and contrast of image data to improve visibility.

[0297] The "inspection result" is a result including a determination of the authenticity of the personal identification document and notification content based on the user's emotional state.

[0298] "Authenticity determination" is the process of determining whether an identification document is genuine or fake.

[0299] This invention combines a system for automatically determining the authenticity of identification documents with an emotion engine that analyzes the user's emotional state. The system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of the data, and compare it with existing databases to determine authenticity. It then recognizes the user's emotional state in real time and optimizes the notification process of the screening results based on the emotional information.

[0300] System configuration and operation overview

[0301] 1. User Actions

[0302] The user accesses the online application system using an internet-connected device (e.g., smartphone, PC) and takes a photo of their identification document (e.g., driver's license, passport) or selects and uploads an existing image file. This action begins the first step in the identity verification process.

[0303] 2. Server Processing

[0304] The server stores the image data received from the user and then pre-processes it, which includes the following steps:

[0305] Resize: Change the image resolution to the specified size.

[0306] Denoising: Removing unnecessary information and dust from an image, for example using a Gaussian filter.

[0307] Deskew: Level out the tilt of the image, for example by using the Hough transform.

[0308] Brightness and Contrast Adjustment: Adjust the image brightness and contrast to the optimum level.

[0309] 3. Use of OCR technology

[0310] The preprocessed image is fed to an OCR engine (e.g., Tesseract OCR) to extract text data, which is then saved to a server as a text file.

[0311] 4. Analysis by AI algorithm

[0312] The extracted text data is input into an AI algorithm (e.g., TensorFlow model) to analyze features such as character arrangement, font, position, size, etc. The analysis results are compared with an existing database on the server to calculate the degree of match.

[0313] 5. Authenticity Verification

[0314] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual review is required, and the manual review team is notified.

[0315] 6. Emotion Analysis Using an Emotion Engine

[0316] The server analyzes the user's emotional state in real time based on camera images, audio data, and timing information of input operations acquired from the user's device. This analysis is performed using an emotion engine (e.g., Affectiva SDK). The results of this analysis are sent to the server and used to notify the user of the review results and for follow-up procedures.

[0317] 7. Notification of the review results

[0318] The server integrates the analysis results with the user's emotional state to generate an optimized notification message. The review results are notified to the user through the application system. The notification includes appropriate feedback that reflects the user's emotional state.

[0319] Specific example of system operation

[0320] When a user applies online for a new mobile phone contract, they upload a picture of their driver's license.

[0321] The server receives the image and stores it in a database.

[0322] Preprocessing involves resizing the image, removing noise, straightening the image, and adjusting the brightness and contrast.

[0323] An OCR engine is used to extract text data, which is then analyzed using AI algorithms.

[0324] The server compares the analysis results with a database and calculates the degree of match.

[0325] Based on the degree of match, the authenticity of the document is determined, and an emotion engine is used to collect and analyze user emotion data.

[0326] Finally, the server integrates the analysis results with the emotional information and sends the results of the screening to the application system as an optimized notification message. The user can then check the results on the application site and receive follow-up instructions if necessary.

[0327] Prompt Sentence Examples

[0328] Here are some example prompts to input to a generative AI model:

[0329] When a user uploads a photo of their driver's license, the system automatically pre-processes the data and uses OCR technology and AI algorithms to extract and analyze text data to determine the authenticity of the ID document. It also analyzes the user's emotional state to optimize the notification process of the review results.

[0330] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0331] Processing Step Description

[0332] Step 1:

[0333] Input: The user uses the device to upload an image file of their identification document (e.g., driver's license).

[0334] Specific operation: A user accesses the online application system, selects an image, and clicks the upload button.

[0335] Output: The uploaded image file is sent to the server.

[0336] Step 2:

[0337] Input: The image file received by the server.

[0338] Specific operation: The server receives the image file and stores it in the database.

[0339] Output: Saved image file.

[0340] Step 3:

[0341] Input: Saved image file.

[0342] What it does: The server resizes the image, changing the resolution to the specified size, then applies noise reduction to remove unnecessary information from the image, deskews the image to make it horizontal, and adjusts the brightness and contrast.

[0343] Output: Preprocessed image files.

[0344] Step 4:

[0345] Input: Preprocessed image files.

[0346] Specific operation: The server inputs the preprocessed image into an OCR engine (e.g., Tesseract OCR) and extracts text data from the image.

[0347] Output: The extracted text data.

[0348] Step 5:

[0349] Input: Extracted text data.

[0350] How it works: The server inputs text data into an AI algorithm (e.g., a TensorFlow model) and analyzes features such as character alignment, font, position, and size.

[0351] Output: Features of the parsed character data.

[0352] Step 6:

[0353] Input: Parsed character data features.

[0354] How it works: The server compares the analysis results with the existing database and calculates the degree of match. In this process, it calculates how closely the character arrangement and font match the existing data.

[0355] Output: Match data.

[0356] Step 7:

[0357] Input: Match data.

[0358] How it works: The server classifies the result as "true," "false," or "suspicious" based on the degree of match. If it is suspicious, it sets a flag indicating that manual review is required.

[0359] Output: Classified judgment result.

[0360] Step 8:

[0361] Input: Camera images and audio data obtained from the user's device, as well as timing information for input operations.

[0362] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's emotional state in real time.

[0363] Output: User's emotional state data.

[0364] Step 9:

[0365] Input: Judgment result and user's emotional state data.

[0366] Specific operation: The server integrates the judgment result and emotional information to generate an optimized notification message.

[0367] Output: Informational message.

[0368] Step 10:

[0369] Input: Notification message.

[0370] Specific operation: The server sends a notification message reflecting the review result and emotional state to the user through the application system.

[0371] Output: The final review result for the user to review.

[0372] Through these steps, the server can determine the authenticity of the identification documents and notify the user of the results of the review, taking into account the user's emotional state, thereby improving the user experience.

[0373] (Application example 2)

[0374] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0375] Food delivery services are required to quickly and accurately determine the authenticity of delivery partners' identification documents. It is also important to improve the quality of service by taking into account the emotional state of delivery personnel and users and responding appropriately. Current systems do not have the functionality to fully resolve these issues, so a mechanism is needed to efficiently increase safety and user satisfaction.

[0376] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0377] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, and means for analyzing the emotional state of the user using emotion analysis technology, which enables the authenticity of the delivery partner's identification documents to be determined and optimal notifications to be given based on the emotional state of the user or delivery person.

[0378] "OCR technology" is an optical character recognition technology that analyzes character information input as an image and reads it as digital text.

[0379] An "AI algorithm" is a type of artificial intelligence technology, and refers to a calculation procedure for analyzing data and performing pattern recognition and predictions.

[0380] "Emotion analysis technology" is a technology that analyzes and judges a user's emotional state based on camera images and audio data.

[0381] "Identification documents" are documents used to prove an individual's identity or qualifications, such as a driver's license or passport.

[0382] "Authentication" refers to the process of determining whether an item or data is genuine or counterfeit.

[0383] A "server" is a computer system that serves the purpose of providing services to other computers on a network.

[0384] "Preprocessing" refers to a series of processing steps performed on raw data, including data cleansing and transformation before analysis or recognition.

[0385] "Resizing" refers to the operation of changing the size of an image or data.

[0386] "Noise reduction" is a technology that removes unnecessary information from image and audio data to improve the quality of the data.

[0387] "Tilt correction" is the process of straightening out images or text that are tilted.

[0388] "Brightness and contrast adjustment" refers to the adjustment process used in image processing to optimize the brightness and contrast of an image.

[0389] A "database" is a system for efficiently managing, searching, and updating large amounts of data.

[0390] "Notification" is the process of conveying information or results to a specific audience.

[0391] This invention is a system that verifies the authenticity of delivery partners' identification documents in food delivery services and analyzes the emotional state of users and delivery partners. This system combines OCR technology, AI algorithms, and emotion analysis technology to provide a safe service that provides high levels of user satisfaction.

[0392] System configuration

[0393] The system consists of devices used by users and delivery partners, a server that performs image processing and data analysis, and a cloud service for emotion analysis.

[0394] Hardware and Software

[0395] Device: A smartphone or tablet that allows users and delivery partners to upload images of their identification documents.

[0396] Server: A computer system that stores data, performs preprocessing, performs AI analysis, and notifies results. For example, cloud services (AWS, Google Cloud Platform) are used.

[0397] OCR technology: Extracts text data from images using Google Cloud Vision API.

[0398] AI algorithm: TensorFlow is used to analyze the characteristics of the extracted text data and determine its authenticity.

[0399] Emotion analysis technology: Uses Microsoft Azure Face API to analyze the emotional state of users and delivery partners.

[0400] Data processing and calculation

[0401] The server receives image data of identification documents uploaded by users and delivery partners, and performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment. Next, text data is extracted from the preprocessed image data using OCR technology using the Google Cloud Vision API. The extracted text data undergoes feature analysis using TensorFlow and is compared with an existing database to determine authenticity. Sentiment analysis technology is also used to analyze the emotional state of users and delivery partners in real time.

[0402] Specific examples

[0403] For example, when a delivery partner registers, they upload an image of their driver's license. This image is sent to a server, where it is pre-processed and text data is extracted using OCR technology. It is then analyzed by an AI algorithm to determine its authenticity. At the same time, the delivery partner's emotional state is analyzed from the camera image, and an appropriate notification message is sent based on that emotional state.

[0404] Prompt Sentence Examples

[0405] "Please upload an image of the delivery partner's verification document. We use OCR technology to extract the text data, and AI will determine its authenticity. Next, we will perform sentiment analysis using the camera image and provide the most appropriate notification."

[0406] In this way, the system combines the authentication of identification documents with the analysis of emotional states, making the delivery partner registration process smooth and safe, while also enabling notifications that take into account the emotional state of users and delivery partners.

[0407] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0408] Step 1:

[0409] The user uploads an image of their identity document from their device.

[0410] Input: Image of identification document

[0411] Specific operation: The user uses a smartphone or tablet to take a photo of their identification document, such as a driver's license or passport, and uploads the image using a dedicated application.

[0412] Step 2:

[0413] The terminal transmits the uploaded image data to the server.

[0414] Input: Image of identification document

[0415] Output: Image data transferred to the server

[0416] Specific operation: The device sends the uploaded image data to the server, which receives the data and stores it in a database.

[0417] Step 3:

[0418] The server performs pre-processing of the image data.

[0419] Input: Uploaded image data

[0420] Output: Preprocessed image data

[0421] Specific operation: The server performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment, thereby obtaining image data suitable for analysis.

[0422] Step 4:

[0423] The server extracts character data from the preprocessed image data using OCR technology.

[0424] Input: Preprocessed image data

[0425] Output: Extracted character data

[0426] How it works: The server uses the Google Cloud Vision API to extract text information from the preprocessed image data, and the extracted text data is stored on the server.

[0427] Step 5:

[0428] The server uses an AI algorithm to analyze the extracted text data.

[0429] Input: Extracted character data

[0430] Output: Analysis results (characteristics of character data)

[0431] Specific operation: The server uses TensorFlow to analyze the extracted character data's characteristics, such as alignment, font, position, and size. The analysis results are stored in a database on the server.

[0432] Step 6:

[0433] The server compares the analysis results with an existing database to determine authenticity.

[0434] Input: Analysis results (characteristics of character data)

[0435] Output: Matching result (judgment result)

[0436] Specific operation: The server compares the analysis results with the existing database and calculates the degree of match. Based on the calculated degree of match, it classifies the result as "true," "false," or "suspicious."

[0437] Step 7:

[0438] The server analyzes the user's emotions using emotion analysis technology.

[0439] Input: User's camera image

[0440] Output: Emotion analysis results

[0441] How it works: The server uses the Microsoft Azure Face API to analyze the user's emotional state from their camera image. The analyzed emotional state is then stored on the server.

[0442] Step 8:

[0443] The server notifies the review results based on the judgment results and emotion analysis results.

[0444] Input: Matching result (judgment result), emotion analysis result

[0445] Output: Notification message

[0446] Specific operation: The server combines the matching results and the sentiment analysis results to generate and send the optimal notification message to the user. For example, it customizes the tone and content of the notification message based on the sentiment analysis results.

[0447] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0448] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0449] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0450] [Second embodiment]

[0451] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0452] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0453] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0454] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0455] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0456] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0457] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0458] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0459] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0460] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0461] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0462] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0463] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0464] System configuration and operation overview

[0465] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[0466] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0467] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The result is also notified to the application system, and the results are shared with the user.

[0468] Specific example of system operation

[0469] 1. User Actions

[0470] A user applies for a new mobile phone contract online.

[0471] The user uploads an image of their driver's license and enters the required personal information.

[0472] 2. Server Processing

[0473] The server receives image data from the user and stores it in a database.

[0474] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0475] The preprocessed image is fed to an OCR engine to extract text data.

[0476] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0477] 3. Judgment and Notification

[0478] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0479] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0480] The server notifies the application system of the judgment result and displays the examination result to the user.

[0481] This system automatically verifies the authenticity of identification documents, enabling early detection of fraudulent applications and improving the accuracy of screening. Furthermore, it is possible to reduce financial risks for companies and significantly improve the efficiency of application processing. The system based on this invention can be easily incorporated into online application systems and is applicable to a wide range of uses.

[0482] The processing flow will be explained below.

[0483] Step 1:

[0484] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[0485] Step 2:

[0486] The server receives the image data sent by the user and stores the received image data in a database.

[0487] Step 3:

[0488] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[0489] Image resizing: Adjust the size so that OCR and AI can process it properly.

[0490] Denoise: Apply filtering to reduce noise in the image.

[0491] Deskew: If the image is tilted, correct the tilt.

[0492] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[0493] Step 4:

[0494] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. The extracted data includes name, address, date of birth, etc. The server saves the extracted text data as a text file.

[0495] Step 5:

[0496] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[0497] Step 6:

[0498] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[0499] Step 7:

[0500] The server performs the following classification based on the comparison results:

[0501] "True": If there is a high degree of match with existing data.

[0502] "False": If the match is very low.

[0503] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[0504] Step 8:

[0505] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[0506] Step 9:

[0507] The application site will be set up so that users can check the results of the screening. Users can check on the website whether their application has been approved or not.

[0508] Example 1

[0509] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0510] Conventional identity verification methods require manual verification of document authenticity, which takes a great deal of time and effort. It also makes it difficult to detect fraudulent applications early, resulting in low accuracy and efficiency of screening. Furthermore, with the spread of online applications, there is a growing need for digitization and automated processing of documents, but current systems are inadequate to meet this demand.

[0511] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0512] In this invention, the server includes: means for extracting character data from identification documents using OCR technology; means for analyzing the characteristics of the extracted character data using an AI algorithm; means for comparing the analyzed character data characteristics with an existing database to determine authenticity; means for notifying the review results based on the results of the review; means for the user to upload an image of the identification document to the online application system using a terminal; means for the server to store and preprocess the received image data; means for providing the preprocessed image to an OCR engine to extract text data; means for analyzing the character arrangement, font, position, size, etc. using an AI algorithm; means for calculating the degree of match based on the analysis results and classifying the result as "true," "false," or "suspicious"; and means for notifying a manual review team if the result is suspicious. This enables automatic and highly accurate determination of the authenticity of identification documents and early detection of fraudulent applications. It also improves the efficiency and accuracy of review and can be easily incorporated into online application systems.

[0513] "OCR technology" is optical character recognition technology, which extracts character data from images.

[0514] "AI algorithm" refers to a calculation method or model that uses artificial intelligence, and is a technology used to analyze and process data.

[0515] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[0516] "Image data" refers to image information stored in digital format, and includes images of identification documents.

[0517] "Preprocessing" refers to processing performed to analyze image data in optimal conditions, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[0518] "Text data" refers to character information extracted from image data and is obtained using OCR technology.

[0519] "Feature analysis" is the process of analyzing various aspects or characteristics of data, including character alignment, font, position, size, etc.

[0520] "Database" refers to a system or collection for systematically managing and storing large amounts of data, in which authentic identity verification data is stored.

[0521] "Concordance" is an index that indicates how closely the analysis results match the records in the database.

[0522] A "judgment result" is a conclusion based on analysis and degree of agreement, and is classified as "true," "false," or "doubtful."

[0523] The "Manual Review Team" is a specialized team that performs detailed human review of suspicious judgment results.

[0524] "Online application system" refers to a system that allows various applications and procedures to be carried out via the Internet.

[0525] A "server" is a computer system that stores, processes, and provides data over a network.

[0526] "Terminal" refers to a device such as a computer or smartphone operated by a user.

[0527] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0528] System configuration

[0529] The main components of the system are:

[0530] 1. User device (PC or smartphone)

[0531] 2. Server

[0532] 3. Online Application System

[0533] 4. OCR engine (e.g. Tesseract OCR)

[0534] 5. AI Algorithms

[0535] 6. Database

[0536] Operation overview

[0537] A user uses a terminal to upload an image of their identification document, such as a driver's license, through an online application system. The server then stores the received image data. The stored image data is preprocessed to resize, remove noise, straighten, and adjust brightness and contrast. This preprocessing is necessary to improve the accuracy of OCR analysis.

[0538] After preprocessing, the image is analyzed by an OCR engine to extract text data. The server saves this text data as a text file. The extracted text data is then input into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0539] Judgment and Notification

[0540] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, it is notified to the manual review team, who then conduct a visual review. Finally, the result is notified to the application system, and the results are shared with the user.

[0541] Specific operation example

[0542] User operation

[0543] A user applies for a new mobile phone contract online.

[0544] The user uploads an image of their driver's license and enters the required personal information.

[0545] Server processing

[0546] The server receives image data from the user and stores it in a database.

[0547] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0548] The preprocessed image is fed to an OCR engine to extract text data.

[0549] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0550] Judgment and Notification

[0551] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0552] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0553] The server notifies the application system of the judgment result and displays the examination result to the user.

[0554] As a concrete example, the following prompt sentence is input to a generative AI model:

[0555] Example prompt sentence:

[0556] Please explain in detail the processing steps of the system for determining the authenticity of ID documents. The user uploads an image of their driver's license, the server performs preprocessing and OCR analysis, and then an AI algorithm analyzes the features to determine authenticity.

[0557] This system automatically determines the authenticity of identification documents, significantly improving the accuracy and efficiency of screening. It also enables early detection of fraudulent applications, reducing financial risks for companies. The system based on this invention can be easily incorporated into existing online application systems and is applicable to a wide range of uses.

[0558] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0559] Step 1:

[0560] User upload of identification documents

[0561] A user uses a terminal (PC or smartphone) to access the online application system and uploads an image of an identification document (e.g., a driver's license).

[0562] Input: Image of user's ID document

[0563] Output: Image data uploaded to the online application system

[0564] Step 2:

[0565] Receiving and storing image data by the server

[0566] The server receives the image data uploaded from the online application system. The received image data is stored on the server, and metadata linked to the user ID and application details is also recorded in the image data.

[0567] Input: Uploaded image data and metadata

[0568] Output: Stored image data and associated metadata

[0569] Step 3:

[0570] Preprocessing of image data by the server

[0571] The server performs pre-processing on the stored image data, which includes the following specific operations:

[0572] Resize: Unify image resolution to improve processing efficiency.

[0573] Noise Reduction: Remove unwanted noise in the image.

[0574] Deskew: Corrects the tilt of the image so that the text is level.

[0575] Brightness and Contrast Adjustment: Properly adjust the brightness and contrast of the image to improve the accuracy of OCR analysis.

[0576] Input: Saved image data

[0577] Output: Preprocessed image data

[0578] Step 4:

[0579] Server-based OCR analysis and text data extraction

[0580] After preprocessing, the image data is input to an OCR engine (e.g., Tesseract OCR) to extract text data from the image. The server saves this text data as a text file.

[0581] Input: Preprocessed image data

[0582] Output: Extracted text data

[0583] Step 5:

[0584] Feature analysis using AI algorithms on the server

[0585] The server inputs the extracted text data into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results provide detailed feature information about the text data.

[0586] Input: Extracted text data

[0587] Output: Analyzed feature information

[0588] Step 6:

[0589] Comparison with database by server and calculation of match

[0590] The server compares the analysis results with an existing database and calculates the degree of match, which determines how closely the text matches the authentic text.

[0591] Input: Analyzed feature information and normal data in the database

[0592] Output: Match rating

[0593] Step 7:

[0594] Classification and notification of judgment results by the server

[0595] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is "suspicious," it sends an alert to the manual review team. Finally, the result is notified to the online application system, which then shares the results with the user.

[0596] Input: Match evaluation result

[0597] Output: Classified judgment results and notification information

[0598] For example, when a user applies for a new mobile phone contract, they upload an image of their driver's license. The server receives the image, performs preprocessing and OCR analysis, and then performs AI feature analysis and comparison with a database. If the AI ​​algorithm determines the image to be "suspicious," an alert is sent to the manual review team. After this process, the user is finally notified of the result.

[0599] (Application example 1)

[0600] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0601] Conventional identity verification systems require manual verification, which requires a great deal of time and resources. They also lack reliable methods for preventing the use of fraudulent identity documents. While electronic payment services require the immediate registration of new accounts, they also require highly accurate verification of the authenticity of identity documents. Therefore, there is a growing need for a system that can efficiently and automatically verify the authenticity of identity documents and immediately notify users of the results.

[0602] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0603] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for determining authenticity by comparing the characteristics of the analyzed character data with an existing database, means for notifying the result of the examination based on the determination result, means for transmitting image data of the identification document from the mobile device to the server at the time of new registration, and means for extracting text data from the preprocessed image data and analyzing the characteristics of the character string using a model. This makes it possible to efficiently and automatically determine the authenticity of identification documents in electronic payment services and immediately notify the user of the result.

[0604] "OCR technology" stands for optical character recognition technology, which is a technology that automatically extracts characters from images.

[0605] An "AI algorithm" is an algorithm that uses artificial intelligence, and is a method of automatically learning and making decisions based on large amounts of data.

[0606] "Identification document" refers to an official document used to verify an individual's identity, such as a driver's license or passport.

[0607] "Text data" refers to text information extracted from an image using OCR technology.

[0608] "Feature analysis" is the process of analyzing the characteristics of extracted character data, such as the arrangement, font, position, and size.

[0609] The "existing database" is a database that stores previously accumulated characteristic data of identification documents.

[0610] "Authenticity determination" is the process of determining whether a document is genuine or fake by comparing the characteristics of the analyzed character data with existing databases.

[0611] "Notification of inspection results" is the process of notifying the user or system of the results of the authenticity determination.

[0612] A "mobile device" is a portable electronic device such as a smartphone or tablet.

[0613] A "server" is a computer system that processes data, analyzes image and text data, and manages the results of judgments.

[0614] "Preprocessing" is the process of resizing, removing noise, correcting tilt, and adjusting brightness and contrast to make image data easier to analyze.

[0615] "Model" means a machine learning model trained by an AI algorithm and used to analyze data.

[0616] This invention is a system that efficiently and automatically determines the authenticity of identity documents in electronic payment services and immediately notifies the user of the results. This system eliminates the need for the conventional manual verification process and enables rapid and accurate identity verification.

[0617] System configuration and operation overview

[0618] 1. User Action:

[0619] When a user creates a new account for an electronic payment service, the user takes a picture of their identification document (e.g., driver's license) using a mobile device such as a smartphone and sends it to a server via an application.

[0620] 2. Server Processing:

[0621] The server receives the image data of the ID document sent by the user and performs preprocessing, including image resizing, noise removal, tilt correction, brightness and contrast adjustment, etc. This is done using OpenCV, an open source image processing library.

[0622] After preprocessing, the image is then processed using the OCR engine Pytesseract to extract text data, which contains the necessary character information and is then analyzed using AI algorithms.

[0623] The extracted character data is fed into a pre-trained machine learning model, which is trained using a machine learning library such as scikit-learn, to analyze specific features of the string (e.g., character alignment, font, position, size, etc.).

[0624] 3. Authentication and Notification:

[0625] The server compares the characteristics of the analyzed text data with an existing database and calculates the degree of match. Based on this degree of match, the authenticity is classified into three categories: "true," "false," or "suspicious." If the authenticity is doubtful, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified.

[0626] The result of the check is immediately notified to the user via the application, for example, a message such as "Your identity verification documents are valid, so your account creation has been completed."

[0627] Specific use cases

[0628] Register a new account:

[0629] A user launches an application to create a new electronic payment account and uploads a photo of their driver's license.

[0630] The server receives the images, pre-processes them, and analyzes them using OCR and AI algorithms.

[0631] The result of the judgment is immediately notified to the user, and if it is true, the account creation is completed.

[0632] Prompt Sentence Examples

[0633] To create a new e-payment account, please upload a photo of your driver's license. The system will automatically verify the authenticity of your ID and notify you of the processing result.

[0634] The system is expected to speed up the process of registering new accounts for electronic payment services, improve the accuracy of identity verification, and significantly improve the user experience.

[0635] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0636] Step 1:

[0637] To create a new electronic payment account, a user takes a photo of their identity document with their mobile device and uploads the image to a server via an application. The input is the image data of the identity document, which is then sent to the server. The output is the image data stored on the server.

[0638] Step 2:

[0639] The server performs preprocessing on the received image data. Specifically, it uses OpenCV to resize, remove noise, correct distortion, and adjust brightness and contrast. The input is the uploaded image data, and the output is the preprocessed image data.

[0640] Step 3:

[0641] The server supplies the preprocessed image data to an OCR engine (Pytesseract) to extract character data. The input is the preprocessed image data, and the output is the extracted character data. At this stage, character information is obtained in text format.

[0642] Step 4:

[0643] The extracted character data is input into an AI algorithm for feature analysis. Specifically, a machine learning library (such as scikit-learn) is used to analyze features such as character arrangement, font, position, and size. The input is text data, and the output is the analyzed feature data.

[0644] Step 5:

[0645] The server compares the analyzed data with the existing database and calculates the degree of match. The input is the analyzed data and the existing database, and the output is a score of the degree of match. In this step, an authentication algorithm based on a specific model is used.

[0646] Step 6:

[0647] Based on the degree of match, the authenticity is judged and the judgment result is classified into three categories: "true," "false," or "suspicious." The input is the match score, and the output is the status of the judgment result. In this step, if the result is suspicious, a flag is set indicating that manual review is required.

[0648] Step 7:

[0649] The server notifies the user of the final review result through the application system. Specifically, the review result is displayed on the user's mobile device. The input is the review result status, and the output is the review result displayed to the user. This result allows the user to confirm whether or not the account creation is complete.

[0650] The above steps effectively automate the new account registration process for electronic payment services.

[0651] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0652] This invention combines an emotion engine with a system for automatically determining the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It then combines an emotion engine that recognizes the user's emotional state and optimizes the notification process of the screening results based on the emotional information.

[0653] System configuration and operation overview

[0654] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[0655] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0656] Based on this degree of match, the server classifies the judgment result as "true," "false," or "suspicious." If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The judgment result is also notified to the application system, and the result is shared with the user. At the same time, the emotion engine analyzes the user's emotional state and reflects this information in the inspection process.

[0657] Overview of Emotion Engine Operation

[0658] The emotion engine analyzes the user's emotional state in real time based on camera images, audio data, timing information of input operations, etc. The results of this analysis are sent to the server and used to notify the review results and for follow-up procedures.

[0659] Specific example of system operation

[0660] 1. User Actions

[0661] A user applies for a new mobile phone contract online.

[0662] The user uploads an image of their driver's license and enters the required personal information.

[0663] 2. Server Processing

[0664] The server receives image data from the user and stores it in a database.

[0665] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0666] The preprocessed image is fed to an OCR engine to extract text data.

[0667] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0668] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0669] 3. Coordination of judgment and emotion analysis

[0670] The server classifies the results as "true," "false," or "suspicious" based on the degree of match.

[0671] The server starts the emotion engine and collects and analyzes the user's emotion data.

[0672] The server integrates the analysis results and sends the review results to the application system. Emotional data is used to customize notification methods and message content.

[0673] 4. Notice to Users

[0674] The user checks the results of the review on the application site, and the notification includes appropriate feedback that reflects the results of the sentiment analysis.

[0675] If necessary, follow-up suggestions are provided to the user.

[0676] This system based on the present invention not only improves the accuracy of determining the authenticity of identification documents, but also improves the user experience and further prevents fraud by enabling flexible notification of review results that take into account the user's emotional state.

[0677] The processing flow will be explained below.

[0678] Step 1:

[0679] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[0680] Step 2:

[0681] The server receives the image data sent by the user and stores the received image data in a database.

[0682] Step 3:

[0683] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[0684] Image resizing: Adjust the size so that OCR and AI can process it properly.

[0685] Denoise: Apply filtering to reduce noise in the image.

[0686] Deskew: If the image is tilted, correct the tilt.

[0687] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[0688] Step 4:

[0689] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. This data includes the name, address, date of birth, etc. The server saves the extracted text data as a text file for further processing.

[0690] Step 5:

[0691] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[0692] Step 6:

[0693] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[0694] Step 7:

[0695] The server performs the following classification based on the comparison results:

[0696] "True": If there is a high degree of match with existing data.

[0697] "False": If the match is very low.

[0698] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[0699] Step 8:

[0700] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[0701] Step 9:

[0702] The server starts the emotion engine and collects camera images, audio data, and timing information of input operations from the user's device. The emotion engine uses this data to analyze the user's emotional state in real time.

[0703] Step 10:

[0704] The server integrates the emotion analysis results into the review result notification process, and optimizes the content and method of review result notification taking into account the user's emotional state.

[0705] Step 11:

[0706] The server sends the final review results, feedback based on the emotional state, and follow-up suggestions, if necessary, to the application system.

[0707] Step 12:

[0708] The user checks the results of the review on the application site. The notification includes appropriate feedback that reflects the results of the sentiment analysis. The user can then check the results and take the next action.

[0709] Example 2

[0710] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0711] In conventional identity verification systems, determining the authenticity of identity documents is often done manually, which is time-consuming and labor-intensive. Even in automated systems, there is no notification process that takes into account the user's emotional state, making it difficult to improve the user experience. Furthermore, the lack of emotion analysis means that notification of results is uniform, making it difficult to respond flexibly.

[0712] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0713] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity, emotion engine means for analyzing the user's emotional state in real time, means for optimizing the notification process of the examination result based on the analysis result of the emotion engine means, and means for notifying the examination result based on the determination result. This improves the accuracy of authenticity determination and enables a flexible notification process based on the user's emotional state, significantly improving the user experience.

[0714] "OCR technology" stands for optical character recognition technology, which recognizes characters from image data and extracts them as text data.

[0715] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[0716] An "AI algorithm" is an algorithm based on artificial intelligence technology that learns, analyzes, and predicts specific tasks.

[0717] An "emotion engine" is a system for analyzing a user's emotional state, and is a technology that recognizes emotions in real time based on input information such as camera images and audio data.

[0718] "Preprocessing" refers to processing performed to prepare image data in a format that is easier to analyze, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[0719] "Resizing" is an operation for changing the resolution and size of image data, and is a process for adjusting the resolution to a specified value.

[0720] "Noise reduction" is a process that removes unnecessary information and dust from image data, and is performed to improve image quality.

[0721] "Tilt correction" is a process of correcting the tilt of image data and leveling it.

[0722] "Brightness and contrast adjustment" is a process for adjusting the brightness and contrast of image data to improve visibility.

[0723] The "inspection result" is a result including a determination of the authenticity of the personal identification document and notification content based on the user's emotional state.

[0724] "Authenticity determination" is the process of determining whether an identification document is genuine or fake.

[0725] This invention combines a system for automatically determining the authenticity of identification documents with an emotion engine that analyzes the user's emotional state. The system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of the data, and compare it with existing databases to determine authenticity. It then recognizes the user's emotional state in real time and optimizes the notification process of the screening results based on the emotional information.

[0726] System configuration and operation overview

[0727] 1. User Actions

[0728] The user accesses the online application system using an internet-connected device (e.g., smartphone, PC) and takes a photo of their identification document (e.g., driver's license, passport) or selects and uploads an existing image file. This action begins the first step in the identity verification process.

[0729] 2. Server Processing

[0730] The server stores the image data received from the user and then pre-processes it, which includes the following steps:

[0731] Resize: Change the image resolution to the specified size.

[0732] Denoising: Removing unnecessary information and dust from an image, for example using a Gaussian filter.

[0733] Deskew: Level out the tilt of the image, for example by using the Hough transform.

[0734] Brightness and Contrast Adjustment: Adjust the image brightness and contrast to the optimum level.

[0735] 3. Use of OCR technology

[0736] The preprocessed image is fed to an OCR engine (e.g., Tesseract OCR) to extract text data, which is then saved to a server as a text file.

[0737] 4. Analysis by AI algorithm

[0738] The extracted text data is input into an AI algorithm (e.g., TensorFlow model) to analyze features such as character arrangement, font, position, size, etc. The analysis results are compared with an existing database on the server to calculate the degree of match.

[0739] 5. Authenticity Verification

[0740] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual review is required, and the manual review team is notified.

[0741] 6. Emotion Analysis Using an Emotion Engine

[0742] The server analyzes the user's emotional state in real time based on camera images, audio data, and timing information of input operations acquired from the user's device. This analysis is performed using an emotion engine (e.g., Affectiva SDK). The results of this analysis are sent to the server and used to notify the user of the review results and for follow-up procedures.

[0743] 7. Notification of the review results

[0744] The server integrates the analysis results with the user's emotional state to generate an optimized notification message. The review results are notified to the user through the application system. The notification includes appropriate feedback that reflects the user's emotional state.

[0745] Specific example of system operation

[0746] When a user applies online for a new mobile phone contract, they upload a picture of their driver's license.

[0747] The server receives the image and stores it in a database.

[0748] Preprocessing involves resizing the image, removing noise, straightening the image, and adjusting the brightness and contrast.

[0749] An OCR engine is used to extract text data, which is then analyzed using AI algorithms.

[0750] The server compares the analysis results with a database and calculates the degree of match.

[0751] Based on the degree of match, the authenticity of the document is determined, and an emotion engine is used to collect and analyze user emotion data.

[0752] Finally, the server integrates the analysis results with the emotional information and sends the results of the screening to the application system as an optimized notification message. The user can then check the results on the application site and receive follow-up instructions if necessary.

[0753] Prompt Sentence Examples

[0754] Here are some example prompts to input to a generative AI model:

[0755] When a user uploads a photo of their driver's license, the system automatically pre-processes the data and uses OCR technology and AI algorithms to extract and analyze text data to determine the authenticity of the ID document. It also analyzes the user's emotional state to optimize the notification process of the review results.

[0756] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0757] Processing Step Description

[0758] Step 1:

[0759] Input: The user uses the device to upload an image file of their identification document (e.g., driver's license).

[0760] Specific operation: A user accesses the online application system, selects an image, and clicks the upload button.

[0761] Output: The uploaded image file is sent to the server.

[0762] Step 2:

[0763] Input: The image file received by the server.

[0764] Specific operation: The server receives the image file and stores it in the database.

[0765] Output: Saved image file.

[0766] Step 3:

[0767] Input: Saved image file.

[0768] What it does: The server resizes the image, changing the resolution to the specified size, then applies noise reduction to remove unnecessary information from the image, deskews the image to make it horizontal, and adjusts the brightness and contrast.

[0769] Output: Preprocessed image files.

[0770] Step 4:

[0771] Input: Preprocessed image files.

[0772] Specific operation: The server inputs the preprocessed image into an OCR engine (e.g., Tesseract OCR) and extracts text data from the image.

[0773] Output: The extracted text data.

[0774] Step 5:

[0775] Input: Extracted text data.

[0776] How it works: The server inputs text data into an AI algorithm (e.g., a TensorFlow model) and analyzes features such as character alignment, font, position, and size.

[0777] Output: Features of the parsed character data.

[0778] Step 6:

[0779] Input: Parsed character data features.

[0780] How it works: The server compares the analysis results with the existing database and calculates the degree of match. In this process, it calculates how closely the character arrangement and font match the existing data.

[0781] Output: Match data.

[0782] Step 7:

[0783] Input: Match data.

[0784] How it works: The server classifies the result as "true," "false," or "suspicious" based on the degree of match. If it is suspicious, it sets a flag indicating that manual review is required.

[0785] Output: Classified judgment result.

[0786] Step 8:

[0787] Input: Camera images and audio data obtained from the user's device, as well as timing information for input operations.

[0788] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's emotional state in real time.

[0789] Output: User's emotional state data.

[0790] Step 9:

[0791] Input: Judgment result and user's emotional state data.

[0792] Specific operation: The server integrates the judgment result and emotional information to generate an optimized notification message.

[0793] Output: Informational message.

[0794] Step 10:

[0795] Input: Notification message.

[0796] Specific operation: The server sends a notification message reflecting the review result and emotional state to the user through the application system.

[0797] Output: The final review result for the user to review.

[0798] Through these steps, the server can determine the authenticity of the identification documents and notify the user of the results of the review, taking into account the user's emotional state, thereby improving the user experience.

[0799] (Application example 2)

[0800] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0801] Food delivery services are required to quickly and accurately determine the authenticity of delivery partners' identification documents. It is also important to improve the quality of service by taking into account the emotional state of delivery personnel and users and responding appropriately. Current systems do not have the functionality to fully resolve these issues, so a mechanism is needed to efficiently increase safety and user satisfaction.

[0802] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0803] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, and means for analyzing the emotional state of the user using emotion analysis technology, which enables the authenticity of the delivery partner's identification documents to be determined and optimal notifications to be given based on the emotional state of the user or delivery person.

[0804] "OCR technology" is an optical character recognition technology that analyzes character information input as an image and reads it as digital text.

[0805] An "AI algorithm" is a type of artificial intelligence technology, and refers to a calculation procedure for analyzing data and performing pattern recognition and predictions.

[0806] "Emotion analysis technology" is a technology that analyzes and judges a user's emotional state based on camera images and audio data.

[0807] "Identification documents" are documents used to prove an individual's identity or qualifications, such as a driver's license or passport.

[0808] "Authentication" refers to the process of determining whether an item or data is genuine or counterfeit.

[0809] A "server" is a computer system that serves the purpose of providing services to other computers on a network.

[0810] "Preprocessing" refers to a series of processing steps performed on raw data, including data cleansing and transformation before analysis or recognition.

[0811] "Resizing" refers to the operation of changing the size of an image or data.

[0812] "Noise reduction" is a technology that removes unnecessary information from image and audio data to improve the quality of the data.

[0813] "Tilt correction" is the process of straightening out images or text that are tilted.

[0814] "Brightness and contrast adjustment" refers to the adjustment process used in image processing to optimize the brightness and contrast of an image.

[0815] A "database" is a system for efficiently managing, searching, and updating large amounts of data.

[0816] "Notification" is the process of conveying information or results to a specific audience.

[0817] This invention is a system that verifies the authenticity of delivery partners' identification documents in food delivery services and analyzes the emotional state of users and delivery partners. This system combines OCR technology, AI algorithms, and emotion analysis technology to provide a safe service that provides high levels of user satisfaction.

[0818] System configuration

[0819] The system consists of devices used by users and delivery partners, a server that performs image processing and data analysis, and a cloud service for emotion analysis.

[0820] Hardware and Software

[0821] Device: A smartphone or tablet that allows users and delivery partners to upload images of their identification documents.

[0822] Server: A computer system that stores data, performs preprocessing, performs AI analysis, and notifies results. For example, cloud services (AWS, Google Cloud Platform) are used.

[0823] OCR technology: Extracts text data from images using Google Cloud Vision API.

[0824] AI algorithm: TensorFlow is used to analyze the characteristics of the extracted text data and determine its authenticity.

[0825] Emotion analysis technology: Uses Microsoft Azure Face API to analyze the emotional state of users and delivery partners.

[0826] Data processing and calculation

[0827] The server receives image data of identification documents uploaded by users and delivery partners, and performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment. Next, text data is extracted from the preprocessed image data using OCR technology using the Google Cloud Vision API. The extracted text data undergoes feature analysis using TensorFlow and is compared with an existing database to determine authenticity. Sentiment analysis technology is also used to analyze the emotional state of users and delivery partners in real time.

[0828] Specific examples

[0829] For example, when a delivery partner registers, they upload an image of their driver's license. This image is sent to a server, where it is pre-processed and text data is extracted using OCR technology. It is then analyzed by an AI algorithm to determine its authenticity. At the same time, the delivery partner's emotional state is analyzed from the camera image, and an appropriate notification message is sent based on that emotional state.

[0830] Prompt Sentence Examples

[0831] "Please upload an image of the delivery partner's verification document. We use OCR technology to extract the text data, and AI will determine its authenticity. Next, we will perform sentiment analysis using the camera image and provide the most appropriate notification."

[0832] In this way, the system combines the authentication of identification documents with the analysis of emotional states, making the delivery partner registration process smooth and safe, while also enabling notifications that take into account the emotional state of users and delivery partners.

[0833] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0834] Step 1:

[0835] The user uploads an image of their identity document from their device.

[0836] Input: Image of identification document

[0837] Specific operation: The user uses a smartphone or tablet to take a photo of their identification document, such as a driver's license or passport, and uploads the image using a dedicated application.

[0838] Step 2:

[0839] The terminal transmits the uploaded image data to the server.

[0840] Input: Image of identification document

[0841] Output: Image data transferred to the server

[0842] Specific operation: The device sends the uploaded image data to the server, which receives the data and stores it in a database.

[0843] Step 3:

[0844] The server performs pre-processing of the image data.

[0845] Input: Uploaded image data

[0846] Output: Preprocessed image data

[0847] Specific operation: The server performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment, thereby obtaining image data suitable for analysis.

[0848] Step 4:

[0849] The server extracts character data from the preprocessed image data using OCR technology.

[0850] Input: Preprocessed image data

[0851] Output: Extracted character data

[0852] How it works: The server uses the Google Cloud Vision API to extract text information from the preprocessed image data, and the extracted text data is stored on the server.

[0853] Step 5:

[0854] The server uses an AI algorithm to analyze the extracted text data.

[0855] Input: Extracted character data

[0856] Output: Analysis results (characteristics of character data)

[0857] Specific operation: The server uses TensorFlow to analyze the extracted character data's characteristics, such as alignment, font, position, and size. The analysis results are stored in a database on the server.

[0858] Step 6:

[0859] The server compares the analysis results with an existing database to determine authenticity.

[0860] Input: Analysis results (characteristics of character data)

[0861] Output: Matching result (judgment result)

[0862] Specific operation: The server compares the analysis results with the existing database and calculates the degree of match. Based on the calculated degree of match, it classifies the result as "true," "false," or "suspicious."

[0863] Step 7:

[0864] The server analyzes the user's emotions using emotion analysis technology.

[0865] Input: User's camera image

[0866] Output: Emotion analysis results

[0867] How it works: The server uses the Microsoft Azure Face API to analyze the user's emotional state from their camera image. The analyzed emotional state is then stored on the server.

[0868] Step 8:

[0869] The server notifies the review results based on the judgment results and emotion analysis results.

[0870] Input: Matching result (judgment result), emotion analysis result

[0871] Output: Notification message

[0872] Specific operation: The server combines the matching results and the sentiment analysis results to generate and send the optimal notification message to the user. For example, it customizes the tone and content of the notification message based on the sentiment analysis results.

[0873] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0874] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0875] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0876] [Third embodiment]

[0877] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0878] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0879] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0880] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0881] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0882] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0883] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0884] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0885] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0886] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0887] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0888] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0889] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0890] System configuration and operation overview

[0891] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[0892] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0893] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The result is also notified to the application system, and the results are shared with the user.

[0894] Specific example of system operation

[0895] 1. User Actions

[0896] A user applies for a new mobile phone contract online.

[0897] The user uploads an image of their driver's license and enters the required personal information.

[0898] 2. Server Processing

[0899] The server receives image data from the user and stores it in a database.

[0900] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0901] The preprocessed image is fed to an OCR engine to extract text data.

[0902] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0903] 3. Judgment and Notification

[0904] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0905] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0906] The server notifies the application system of the judgment result and displays the examination result to the user.

[0907] This system automatically verifies the authenticity of identification documents, enabling early detection of fraudulent applications and improving the accuracy of screening. Furthermore, it is possible to reduce financial risks for companies and significantly improve the efficiency of application processing. The system based on this invention can be easily incorporated into online application systems and is applicable to a wide range of uses.

[0908] The processing flow will be explained below.

[0909] Step 1:

[0910] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[0911] Step 2:

[0912] The server receives the image data sent by the user and stores the received image data in a database.

[0913] Step 3:

[0914] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[0915] Image resizing: Adjust the size so that OCR and AI can process it properly.

[0916] Denoise: Apply filtering to reduce noise in the image.

[0917] Deskew: If the image is tilted, correct the tilt.

[0918] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[0919] Step 4:

[0920] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. The extracted data includes name, address, date of birth, etc. The server saves the extracted text data as a text file.

[0921] Step 5:

[0922] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[0923] Step 6:

[0924] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[0925] Step 7:

[0926] The server performs the following classification based on the comparison results:

[0927] "True": If there is a high degree of match with existing data.

[0928] "False": If the match is very low.

[0929] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[0930] Step 8:

[0931] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[0932] Step 9:

[0933] The application site will be set up so that users can check the results of the screening. Users can check on the website whether their application has been approved or not.

[0934] Example 1

[0935] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0936] Conventional identity verification methods require manual verification of document authenticity, which takes a great deal of time and effort. It also makes it difficult to detect fraudulent applications early, resulting in low accuracy and efficiency of screening. Furthermore, with the spread of online applications, there is a growing need for digitization and automated processing of documents, but current systems are inadequate to meet this demand.

[0937] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0938] In this invention, the server includes: means for extracting character data from identification documents using OCR technology; means for analyzing the characteristics of the extracted character data using an AI algorithm; means for comparing the analyzed character data characteristics with an existing database to determine authenticity; means for notifying the review results based on the results of the review; means for the user to upload an image of the identification document to the online application system using a terminal; means for the server to store and preprocess the received image data; means for providing the preprocessed image to an OCR engine to extract text data; means for analyzing the character arrangement, font, position, size, etc. using an AI algorithm; means for calculating the degree of match based on the analysis results and classifying the result as "true," "false," or "suspicious"; and means for notifying a manual review team if the result is suspicious. This enables automatic and highly accurate determination of the authenticity of identification documents and early detection of fraudulent applications. It also improves the efficiency and accuracy of review and can be easily incorporated into online application systems.

[0939] "OCR technology" is optical character recognition technology, which extracts character data from images.

[0940] "AI algorithm" refers to a calculation method or model that uses artificial intelligence, and is a technology used to analyze and process data.

[0941] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[0942] "Image data" refers to image information stored in digital format, and includes images of identification documents.

[0943] "Preprocessing" refers to processing performed to analyze image data in optimal conditions, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[0944] "Text data" refers to character information extracted from image data and is obtained using OCR technology.

[0945] "Feature analysis" is the process of analyzing various aspects or characteristics of data, including character alignment, font, position, size, etc.

[0946] "Database" refers to a system or collection for systematically managing and storing large amounts of data, in which authentic identity verification data is stored.

[0947] "Concordance" is an index that indicates how closely the analysis results match the records in the database.

[0948] A "judgment result" is a conclusion based on analysis and degree of agreement, and is classified as "true," "false," or "doubtful."

[0949] The "Manual Review Team" is a specialized team that performs detailed human review of suspicious judgment results.

[0950] "Online application system" refers to a system that allows various applications and procedures to be carried out via the Internet.

[0951] A "server" is a computer system that stores, processes, and provides data over a network.

[0952] "Terminal" refers to a device such as a computer or smartphone operated by a user.

[0953] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[0954] System configuration

[0955] The main components of the system are:

[0956] 1. User device (PC or smartphone)

[0957] 2. Server

[0958] 3. Online Application System

[0959] 4. OCR engine (e.g. Tesseract OCR)

[0960] 5. AI Algorithms

[0961] 6. Database

[0962] Operation overview

[0963] A user uses a terminal to upload an image of their identification document, such as a driver's license, through an online application system. The server then stores the received image data. The stored image data is preprocessed to resize, remove noise, straighten, and adjust brightness and contrast. This preprocessing is necessary to improve the accuracy of OCR analysis.

[0964] After preprocessing, the image is analyzed by an OCR engine to extract text data. The server saves this text data as a text file. The extracted text data is then input into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[0965] Judgment and Notification

[0966] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, it is notified to the manual review team, who then conduct a visual review. Finally, the result is notified to the application system, and the results are shared with the user.

[0967] Specific operation example

[0968] User operation

[0969] A user applies for a new mobile phone contract online.

[0970] The user uploads an image of their driver's license and enters the required personal information.

[0971] Server processing

[0972] The server receives image data from the user and stores it in a database.

[0973] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[0974] The preprocessed image is fed to an OCR engine to extract text data.

[0975] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[0976] Judgment and Notification

[0977] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[0978] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[0979] The server notifies the application system of the judgment result and displays the examination result to the user.

[0980] As a concrete example, the following prompt sentence is input to a generative AI model:

[0981] Example prompt sentence:

[0982] Please explain in detail the processing steps of the system for determining the authenticity of ID documents. The user uploads an image of their driver's license, the server performs preprocessing and OCR analysis, and then an AI algorithm analyzes the features to determine authenticity.

[0983] This system automatically determines the authenticity of identification documents, significantly improving the accuracy and efficiency of screening. It also enables early detection of fraudulent applications, reducing financial risks for companies. The system based on this invention can be easily incorporated into existing online application systems and is applicable to a wide range of uses.

[0984] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0985] Step 1:

[0986] User upload of identification documents

[0987] A user uses a terminal (PC or smartphone) to access the online application system and uploads an image of an identification document (e.g., a driver's license).

[0988] Input: Image of user's ID document

[0989] Output: Image data uploaded to the online application system

[0990] Step 2:

[0991] Receiving and storing image data by the server

[0992] The server receives the image data uploaded from the online application system. The received image data is stored on the server, and metadata linked to the user ID and application details is also recorded in the image data.

[0993] Input: Uploaded image data and metadata

[0994] Output: Stored image data and associated metadata

[0995] Step 3:

[0996] Preprocessing of image data by the server

[0997] The server performs pre-processing on the stored image data, which includes the following specific operations:

[0998] Resize: Unify image resolution to improve processing efficiency.

[0999] Noise Reduction: Remove unwanted noise in the image.

[1000] Deskew: Corrects the tilt of the image so that the text is level.

[1001] Brightness and Contrast Adjustment: Properly adjust the brightness and contrast of the image to improve the accuracy of OCR analysis.

[1002] Input: Saved image data

[1003] Output: Preprocessed image data

[1004] Step 4:

[1005] Server-based OCR analysis and text data extraction

[1006] After preprocessing, the image data is input to an OCR engine (e.g., Tesseract OCR) to extract text data from the image. The server saves this text data as a text file.

[1007] Input: Preprocessed image data

[1008] Output: Extracted text data

[1009] Step 5:

[1010] Feature analysis using AI algorithms on the server

[1011] The server inputs the extracted text data into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results provide detailed feature information about the text data.

[1012] Input: Extracted text data

[1013] Output: Analyzed feature information

[1014] Step 6:

[1015] Comparison with database by server and calculation of match

[1016] The server compares the analysis results with an existing database and calculates the degree of match, which determines how closely the text matches the authentic text.

[1017] Input: Analyzed feature information and normal data in the database

[1018] Output: Match rating

[1019] Step 7:

[1020] Classification and notification of judgment results by the server

[1021] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is "suspicious," it sends an alert to the manual review team. Finally, the result is notified to the online application system, which then shares the results with the user.

[1022] Input: Match evaluation result

[1023] Output: Classified judgment results and notification information

[1024] For example, when a user applies for a new mobile phone contract, they upload an image of their driver's license. The server receives the image, performs preprocessing and OCR analysis, and then performs AI feature analysis and comparison with a database. If the AI ​​algorithm determines the image to be "suspicious," an alert is sent to the manual review team. After this process, the user is finally notified of the result.

[1025] (Application example 1)

[1026] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1027] Conventional identity verification systems require manual verification, which requires a great deal of time and resources. They also lack reliable methods for preventing the use of fraudulent identity documents. While electronic payment services require the immediate registration of new accounts, they also require highly accurate verification of the authenticity of identity documents. Therefore, there is a growing need for a system that can efficiently and automatically verify the authenticity of identity documents and immediately notify users of the results.

[1028] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1029] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for determining authenticity by comparing the characteristics of the analyzed character data with an existing database, means for notifying the result of the examination based on the determination result, means for transmitting image data of the identification document from the mobile device to the server at the time of new registration, and means for extracting text data from the preprocessed image data and analyzing the characteristics of the character string using a model. This makes it possible to efficiently and automatically determine the authenticity of identification documents in electronic payment services and immediately notify the user of the result.

[1030] "OCR technology" stands for optical character recognition technology, which is a technology that automatically extracts characters from images.

[1031] An "AI algorithm" is an algorithm that uses artificial intelligence, and is a method of automatically learning and making decisions based on large amounts of data.

[1032] "Identification document" refers to an official document used to verify an individual's identity, such as a driver's license or passport.

[1033] "Text data" refers to text information extracted from an image using OCR technology.

[1034] "Feature analysis" is the process of analyzing the characteristics of extracted character data, such as the arrangement, font, position, and size.

[1035] The "existing database" is a database that stores previously accumulated characteristic data of identification documents.

[1036] "Authenticity determination" is the process of determining whether a document is genuine or fake by comparing the characteristics of the analyzed character data with existing databases.

[1037] "Notification of inspection results" is the process of notifying the user or system of the results of the authenticity determination.

[1038] A "mobile device" is a portable electronic device such as a smartphone or tablet.

[1039] A "server" is a computer system that processes data, analyzes image and text data, and manages the results of judgments.

[1040] "Preprocessing" is the process of resizing, removing noise, correcting tilt, and adjusting brightness and contrast to make image data easier to analyze.

[1041] "Model" means a machine learning model trained by an AI algorithm and used to analyze data.

[1042] This invention is a system that efficiently and automatically determines the authenticity of identity documents in electronic payment services and immediately notifies the user of the results. This system eliminates the need for the conventional manual verification process and enables rapid and accurate identity verification.

[1043] System configuration and operation overview

[1044] 1. User Action:

[1045] When a user creates a new account for an electronic payment service, the user takes a picture of their identification document (e.g., driver's license) using a mobile device such as a smartphone and sends it to a server via an application.

[1046] 2. Server Processing:

[1047] The server receives the image data of the ID document sent by the user and performs preprocessing, including image resizing, noise removal, tilt correction, brightness and contrast adjustment, etc. This is done using OpenCV, an open source image processing library.

[1048] After preprocessing, the image is then processed using the OCR engine Pytesseract to extract text data, which contains the necessary character information and is then analyzed using AI algorithms.

[1049] The extracted character data is fed into a pre-trained machine learning model, which is trained using a machine learning library such as scikit-learn, to analyze specific features of the string (e.g., character alignment, font, position, size, etc.).

[1050] 3. Authentication and Notification:

[1051] The server compares the characteristics of the analyzed text data with an existing database and calculates the degree of match. Based on this degree of match, the authenticity is classified into three categories: "true," "false," or "suspicious." If the authenticity is doubtful, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified.

[1052] The result of the check is immediately notified to the user via the application, for example, a message such as "Your identity verification documents are valid, so your account creation has been completed."

[1053] Specific use cases

[1054] Register a new account:

[1055] A user launches an application to create a new electronic payment account and uploads a photo of their driver's license.

[1056] The server receives the images, pre-processes them, and analyzes them using OCR and AI algorithms.

[1057] The result of the judgment is immediately notified to the user, and if it is true, the account creation is completed.

[1058] Prompt Sentence Examples

[1059] To create a new e-payment account, please upload a photo of your driver's license. The system will automatically verify the authenticity of your ID and notify you of the processing result.

[1060] The system is expected to speed up the process of registering new accounts for electronic payment services, improve the accuracy of identity verification, and significantly improve the user experience.

[1061] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1062] Step 1:

[1063] To create a new electronic payment account, a user takes a photo of their identity document with their mobile device and uploads the image to a server via an application. The input is the image data of the identity document, which is then sent to the server. The output is the image data stored on the server.

[1064] Step 2:

[1065] The server performs preprocessing on the received image data. Specifically, it uses OpenCV to resize, remove noise, correct distortion, and adjust brightness and contrast. The input is the uploaded image data, and the output is the preprocessed image data.

[1066] Step 3:

[1067] The server supplies the preprocessed image data to an OCR engine (Pytesseract) to extract character data. The input is the preprocessed image data, and the output is the extracted character data. At this stage, character information is obtained in text format.

[1068] Step 4:

[1069] The extracted character data is input into an AI algorithm for feature analysis. Specifically, a machine learning library (such as scikit-learn) is used to analyze features such as character arrangement, font, position, and size. The input is text data, and the output is the analyzed feature data.

[1070] Step 5:

[1071] The server compares the analyzed data with the existing database and calculates the degree of match. The input is the analyzed data and the existing database, and the output is a score of the degree of match. In this step, an authentication algorithm based on a specific model is used.

[1072] Step 6:

[1073] Based on the degree of match, the authenticity is judged and the judgment result is classified into three categories: "true," "false," or "suspicious." The input is the match score, and the output is the status of the judgment result. In this step, if the result is suspicious, a flag is set indicating that manual review is required.

[1074] Step 7:

[1075] The server notifies the user of the final review result through the application system. Specifically, the review result is displayed on the user's mobile device. The input is the review result status, and the output is the review result displayed to the user. This result allows the user to confirm whether or not the account creation is complete.

[1076] The above steps effectively automate the new account registration process for electronic payment services.

[1077] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1078] This invention combines an emotion engine with a system for automatically determining the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It then combines an emotion engine that recognizes the user's emotional state and optimizes the notification process of the screening results based on the emotional information.

[1079] System configuration and operation overview

[1080] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[1081] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[1082] Based on this degree of match, the server classifies the judgment result as "true," "false," or "suspicious." If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The judgment result is also notified to the application system, and the result is shared with the user. At the same time, the emotion engine analyzes the user's emotional state and reflects this information in the inspection process.

[1083] Overview of Emotion Engine Operation

[1084] The emotion engine analyzes the user's emotional state in real time based on camera images, audio data, timing information of input operations, etc. The results of this analysis are sent to the server and used to notify the review results and for follow-up procedures.

[1085] Specific example of system operation

[1086] 1. User Actions

[1087] A user applies for a new mobile phone contract online.

[1088] The user uploads an image of their driver's license and enters the required personal information.

[1089] 2. Server Processing

[1090] The server receives image data from the user and stores it in a database.

[1091] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[1092] The preprocessed image is fed to an OCR engine to extract text data.

[1093] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[1094] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[1095] 3. Coordination of judgment and emotion analysis

[1096] The server classifies the results as "true," "false," or "suspicious" based on the degree of match.

[1097] The server starts the emotion engine and collects and analyzes the user's emotion data.

[1098] The server integrates the analysis results and sends the review results to the application system. Emotional data is used to customize notification methods and message content.

[1099] 4. Notice to Users

[1100] The user checks the results of the review on the application site, and the notification includes appropriate feedback that reflects the results of the sentiment analysis.

[1101] If necessary, follow-up suggestions are provided to the user.

[1102] This system based on the present invention not only improves the accuracy of determining the authenticity of identification documents, but also improves the user experience and further prevents fraud by enabling flexible notification of review results that take into account the user's emotional state.

[1103] The processing flow will be explained below.

[1104] Step 1:

[1105] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[1106] Step 2:

[1107] The server receives the image data sent by the user and stores the received image data in a database.

[1108] Step 3:

[1109] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[1110] Image resizing: Adjust the size so that OCR and AI can process it properly.

[1111] Denoise: Apply filtering to reduce noise in the image.

[1112] Deskew: If the image is tilted, correct the tilt.

[1113] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[1114] Step 4:

[1115] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. This data includes the name, address, date of birth, etc. The server saves the extracted text data as a text file for further processing.

[1116] Step 5:

[1117] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[1118] Step 6:

[1119] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[1120] Step 7:

[1121] The server performs the following classification based on the comparison results:

[1122] "True": If there is a high degree of match with existing data.

[1123] "False": If the match is very low.

[1124] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[1125] Step 8:

[1126] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[1127] Step 9:

[1128] The server starts the emotion engine and collects camera images, audio data, and timing information of input operations from the user's device. The emotion engine uses this data to analyze the user's emotional state in real time.

[1129] Step 10:

[1130] The server integrates the emotion analysis results into the review result notification process, and optimizes the content and method of review result notification taking into account the user's emotional state.

[1131] Step 11:

[1132] The server sends the final review results, feedback based on the emotional state, and follow-up suggestions, if necessary, to the application system.

[1133] Step 12:

[1134] The user checks the results of the review on the application site. The notification includes appropriate feedback that reflects the results of the sentiment analysis. The user can then check the results and take the next action.

[1135] Example 2

[1136] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1137] In conventional identity verification systems, determining the authenticity of identity documents is often done manually, which is time-consuming and labor-intensive. Even in automated systems, there is no notification process that takes into account the user's emotional state, making it difficult to improve the user experience. Furthermore, the lack of emotion analysis means that notification of results is uniform, making it difficult to respond flexibly.

[1138] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1139] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity, emotion engine means for analyzing the user's emotional state in real time, means for optimizing the notification process of the examination result based on the analysis result of the emotion engine means, and means for notifying the examination result based on the determination result. This improves the accuracy of authenticity determination and enables a flexible notification process based on the user's emotional state, significantly improving the user experience.

[1140] "OCR technology" stands for optical character recognition technology, which recognizes characters from image data and extracts them as text data.

[1141] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[1142] An "AI algorithm" is an algorithm based on artificial intelligence technology that learns, analyzes, and predicts specific tasks.

[1143] An "emotion engine" is a system for analyzing a user's emotional state, and is a technology that recognizes emotions in real time based on input information such as camera images and audio data.

[1144] "Preprocessing" refers to processing performed to prepare image data in a format that is easier to analyze, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[1145] "Resizing" is an operation for changing the resolution and size of image data, and is a process for adjusting the resolution to a specified value.

[1146] "Noise reduction" is a process that removes unnecessary information and dust from image data, and is performed to improve image quality.

[1147] "Tilt correction" is a process of correcting the tilt of image data and leveling it.

[1148] "Brightness and contrast adjustment" is a process for adjusting the brightness and contrast of image data to improve visibility.

[1149] The "inspection result" is a result including a determination of the authenticity of the personal identification document and notification content based on the user's emotional state.

[1150] "Authenticity determination" is the process of determining whether an identification document is genuine or fake.

[1151] This invention combines a system for automatically determining the authenticity of identification documents with an emotion engine that analyzes the user's emotional state. The system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of the data, and compare it with existing databases to determine authenticity. It then recognizes the user's emotional state in real time and optimizes the notification process of the screening results based on the emotional information.

[1152] System configuration and operation overview

[1153] 1. User Actions

[1154] The user accesses the online application system using an internet-connected device (e.g., smartphone, PC) and takes a photo of their identification document (e.g., driver's license, passport) or selects and uploads an existing image file. This action begins the first step in the identity verification process.

[1155] 2. Server Processing

[1156] The server stores the image data received from the user and then pre-processes it, which includes the following steps:

[1157] Resize: Change the image resolution to the specified size.

[1158] Denoising: Removing unnecessary information and dust from an image, for example using a Gaussian filter.

[1159] Deskew: Level out the tilt of the image, for example by using the Hough transform.

[1160] Brightness and Contrast Adjustment: Adjust the image brightness and contrast to the optimum level.

[1161] 3. Use of OCR technology

[1162] The preprocessed image is fed to an OCR engine (e.g., Tesseract OCR) to extract text data, which is then saved to a server as a text file.

[1163] 4. Analysis by AI algorithm

[1164] The extracted text data is input into an AI algorithm (e.g., TensorFlow model) to analyze features such as character arrangement, font, position, size, etc. The analysis results are compared with an existing database on the server to calculate the degree of match.

[1165] 5. Authenticity Verification

[1166] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual review is required, and the manual review team is notified.

[1167] 6. Emotion Analysis Using an Emotion Engine

[1168] The server analyzes the user's emotional state in real time based on camera images, audio data, and timing information of input operations acquired from the user's device. This analysis is performed using an emotion engine (e.g., Affectiva SDK). The results of this analysis are sent to the server and used to notify the user of the review results and for follow-up procedures.

[1169] 7. Notification of the review results

[1170] The server integrates the analysis results with the user's emotional state to generate an optimized notification message. The review results are notified to the user through the application system. The notification includes appropriate feedback that reflects the user's emotional state.

[1171] Specific example of system operation

[1172] When a user applies online for a new mobile phone contract, they upload a picture of their driver's license.

[1173] The server receives the image and stores it in a database.

[1174] Preprocessing involves resizing the image, removing noise, straightening the image, and adjusting the brightness and contrast.

[1175] An OCR engine is used to extract text data, which is then analyzed using AI algorithms.

[1176] The server compares the analysis results with a database and calculates the degree of match.

[1177] Based on the degree of match, the authenticity of the document is determined, and an emotion engine is used to collect and analyze user emotion data.

[1178] Finally, the server integrates the analysis results with the emotional information and sends the results of the screening to the application system as an optimized notification message. The user can then check the results on the application site and receive follow-up instructions if necessary.

[1179] Prompt Sentence Examples

[1180] Here are some example prompts to input to a generative AI model:

[1181] When a user uploads a photo of their driver's license, the system automatically pre-processes the data and uses OCR technology and AI algorithms to extract and analyze text data to determine the authenticity of the ID document. It also analyzes the user's emotional state to optimize the notification process of the review results.

[1182] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1183] Processing Step Description

[1184] Step 1:

[1185] Input: The user uses the device to upload an image file of their identification document (e.g., driver's license).

[1186] Specific operation: A user accesses the online application system, selects an image, and clicks the upload button.

[1187] Output: The uploaded image file is sent to the server.

[1188] Step 2:

[1189] Input: The image file received by the server.

[1190] Specific operation: The server receives the image file and stores it in the database.

[1191] Output: Saved image file.

[1192] Step 3:

[1193] Input: Saved image file.

[1194] What it does: The server resizes the image, changing the resolution to the specified size, then applies noise reduction to remove unnecessary information from the image, deskews the image to make it horizontal, and adjusts the brightness and contrast.

[1195] Output: Preprocessed image files.

[1196] Step 4:

[1197] Input: Preprocessed image files.

[1198] Specific operation: The server inputs the preprocessed image into an OCR engine (e.g., Tesseract OCR) and extracts text data from the image.

[1199] Output: The extracted text data.

[1200] Step 5:

[1201] Input: Extracted text data.

[1202] How it works: The server inputs text data into an AI algorithm (e.g., a TensorFlow model) and analyzes features such as character alignment, font, position, and size.

[1203] Output: Features of the parsed character data.

[1204] Step 6:

[1205] Input: Parsed character data features.

[1206] How it works: The server compares the analysis results with the existing database and calculates the degree of match. In this process, it calculates how closely the character arrangement and font match the existing data.

[1207] Output: Match data.

[1208] Step 7:

[1209] Input: Match data.

[1210] How it works: The server classifies the result as "true," "false," or "suspicious" based on the degree of match. If it is suspicious, it sets a flag indicating that manual review is required.

[1211] Output: Classified judgment result.

[1212] Step 8:

[1213] Input: Camera images and audio data obtained from the user's device, as well as timing information for input operations.

[1214] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's emotional state in real time.

[1215] Output: User's emotional state data.

[1216] Step 9:

[1217] Input: Judgment result and user's emotional state data.

[1218] Specific operation: The server integrates the judgment result and emotional information to generate an optimized notification message.

[1219] Output: Informational message.

[1220] Step 10:

[1221] Input: Notification message.

[1222] Specific operation: The server sends a notification message reflecting the review result and emotional state to the user through the application system.

[1223] Output: The final review result for the user to review.

[1224] Through these steps, the server can determine the authenticity of the identification documents and notify the user of the results of the review, taking into account the user's emotional state, thereby improving the user experience.

[1225] (Application example 2)

[1226] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1227] Food delivery services are required to quickly and accurately determine the authenticity of delivery partners' identification documents. It is also important to improve the quality of service by taking into account the emotional state of delivery personnel and users and responding appropriately. Current systems do not have the functionality to fully resolve these issues, so a mechanism is needed to efficiently increase safety and user satisfaction.

[1228] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1229] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, and means for analyzing the emotional state of the user using emotion analysis technology, which enables the authenticity of the delivery partner's identification documents to be determined and optimal notifications to be given based on the emotional state of the user or delivery person.

[1230] "OCR technology" is an optical character recognition technology that analyzes character information input as an image and reads it as digital text.

[1231] An "AI algorithm" is a type of artificial intelligence technology, and refers to a calculation procedure for analyzing data and performing pattern recognition and predictions.

[1232] "Emotion analysis technology" is a technology that analyzes and judges a user's emotional state based on camera images and audio data.

[1233] "Identification documents" are documents used to prove an individual's identity or qualifications, such as a driver's license or passport.

[1234] "Authentication" refers to the process of determining whether an item or data is genuine or counterfeit.

[1235] A "server" is a computer system that serves the purpose of providing services to other computers on a network.

[1236] "Preprocessing" refers to a series of processing steps performed on raw data, including data cleansing and transformation before analysis or recognition.

[1237] "Resizing" refers to the operation of changing the size of an image or data.

[1238] "Noise reduction" is a technology that removes unnecessary information from image and audio data to improve the quality of the data.

[1239] "Tilt correction" is the process of straightening out images or text that are tilted.

[1240] "Brightness and contrast adjustment" refers to the adjustment process used in image processing to optimize the brightness and contrast of an image.

[1241] A "database" is a system for efficiently managing, searching, and updating large amounts of data.

[1242] "Notification" is the process of conveying information or results to a specific audience.

[1243] This invention is a system that verifies the authenticity of delivery partners' identification documents in food delivery services and analyzes the emotional state of users and delivery partners. This system combines OCR technology, AI algorithms, and emotion analysis technology to provide a safe service that provides high levels of user satisfaction.

[1244] System configuration

[1245] The system consists of devices used by users and delivery partners, a server that performs image processing and data analysis, and a cloud service for emotion analysis.

[1246] Hardware and Software

[1247] Device: A smartphone or tablet that allows users and delivery partners to upload images of their identification documents.

[1248] Server: A computer system that stores data, performs preprocessing, performs AI analysis, and notifies results. For example, cloud services (AWS, Google Cloud Platform) are used.

[1249] OCR technology: Extracts text data from images using Google Cloud Vision API.

[1250] AI algorithm: TensorFlow is used to analyze the characteristics of the extracted text data and determine its authenticity.

[1251] Emotion analysis technology: Uses Microsoft Azure Face API to analyze the emotional state of users and delivery partners.

[1252] Data processing and calculation

[1253] The server receives image data of identification documents uploaded by users and delivery partners, and performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment. Next, text data is extracted from the preprocessed image data using OCR technology using the Google Cloud Vision API. The extracted text data undergoes feature analysis using TensorFlow and is compared with an existing database to determine authenticity. Sentiment analysis technology is also used to analyze the emotional state of users and delivery partners in real time.

[1254] Specific examples

[1255] For example, when a delivery partner registers, they upload an image of their driver's license. This image is sent to a server, where it is pre-processed and text data is extracted using OCR technology. It is then analyzed by an AI algorithm to determine its authenticity. At the same time, the delivery partner's emotional state is analyzed from the camera image, and an appropriate notification message is sent based on that emotional state.

[1256] Prompt Sentence Examples

[1257] "Please upload an image of the delivery partner's verification document. We use OCR technology to extract the text data, and AI will determine its authenticity. Next, we will perform sentiment analysis using the camera image and provide the most appropriate notification."

[1258] In this way, the system combines the authentication of identification documents with the analysis of emotional states, making the delivery partner registration process smooth and safe, while also enabling notifications that take into account the emotional state of users and delivery partners.

[1259] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1260] Step 1:

[1261] The user uploads an image of their identity document from their device.

[1262] Input: Image of identification document

[1263] Specific operation: The user uses a smartphone or tablet to take a photo of their identification document, such as a driver's license or passport, and uploads the image using a dedicated application.

[1264] Step 2:

[1265] The terminal transmits the uploaded image data to the server.

[1266] Input: Image of identification document

[1267] Output: Image data transferred to the server

[1268] Specific operation: The device sends the uploaded image data to the server, which receives the data and stores it in a database.

[1269] Step 3:

[1270] The server performs pre-processing of the image data.

[1271] Input: Uploaded image data

[1272] Output: Preprocessed image data

[1273] Specific operation: The server performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment, thereby obtaining image data suitable for analysis.

[1274] Step 4:

[1275] The server extracts character data from the preprocessed image data using OCR technology.

[1276] Input: Preprocessed image data

[1277] Output: Extracted character data

[1278] How it works: The server uses the Google Cloud Vision API to extract text information from the preprocessed image data, and the extracted text data is stored on the server.

[1279] Step 5:

[1280] The server uses an AI algorithm to analyze the extracted text data.

[1281] Input: Extracted character data

[1282] Output: Analysis results (characteristics of character data)

[1283] Specific operation: The server uses TensorFlow to analyze the extracted character data's characteristics, such as alignment, font, position, and size. The analysis results are stored in a database on the server.

[1284] Step 6:

[1285] The server compares the analysis results with an existing database to determine authenticity.

[1286] Input: Analysis results (characteristics of character data)

[1287] Output: Matching result (judgment result)

[1288] Specific operation: The server compares the analysis results with the existing database and calculates the degree of match. Based on the calculated degree of match, it classifies the result as "true," "false," or "suspicious."

[1289] Step 7:

[1290] The server analyzes the user's emotions using emotion analysis technology.

[1291] Input: User's camera image

[1292] Output: Emotion analysis results

[1293] How it works: The server uses the Microsoft Azure Face API to analyze the user's emotional state from their camera image. The analyzed emotional state is then stored on the server.

[1294] Step 8:

[1295] The server notifies the review results based on the judgment results and emotion analysis results.

[1296] Input: Matching result (judgment result), emotion analysis result

[1297] Output: Notification message

[1298] Specific operation: The server combines the matching results and the sentiment analysis results to generate and send the optimal notification message to the user. For example, it customizes the tone and content of the notification message based on the sentiment analysis results.

[1299] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1300] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1301] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1302] [Fourth embodiment]

[1303] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1304] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1305] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1306] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1307] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1308] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1309] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1310] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1311] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1312] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1313] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1314] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1315] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1316] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[1317] System configuration and operation overview

[1318] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[1319] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[1320] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The result is also notified to the application system, and the results are shared with the user.

[1321] Specific example of system operation

[1322] 1. User Actions

[1323] A user applies for a new mobile phone contract online.

[1324] The user uploads an image of their driver's license and enters the required personal information.

[1325] 2. Server Processing

[1326] The server receives image data from the user and stores it in a database.

[1327] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[1328] The preprocessed image is fed to an OCR engine to extract text data.

[1329] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[1330] 3. Judgment and Notification

[1331] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[1332] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[1333] The server notifies the application system of the judgment result and displays the examination result to the user.

[1334] This system automatically verifies the authenticity of identification documents, enabling early detection of fraudulent applications and improving the accuracy of screening. Furthermore, it is possible to reduce financial risks for companies and significantly improve the efficiency of application processing. The system based on this invention can be easily incorporated into online application systems and is applicable to a wide range of uses.

[1335] The processing flow will be explained below.

[1336] Step 1:

[1337] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[1338] Step 2:

[1339] The server receives the image data sent by the user and stores the received image data in a database.

[1340] Step 3:

[1341] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[1342] Image resizing: Adjust the size so that OCR and AI can process it properly.

[1343] Denoise: Apply filtering to reduce noise in the image.

[1344] Deskew: If the image is tilted, correct the tilt.

[1345] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[1346] Step 4:

[1347] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. The extracted data includes name, address, date of birth, etc. The server saves the extracted text data as a text file.

[1348] Step 5:

[1349] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[1350] Step 6:

[1351] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[1352] Step 7:

[1353] The server performs the following classification based on the comparison results:

[1354] "True": If there is a high degree of match with existing data.

[1355] "False": If the match is very low.

[1356] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[1357] Step 8:

[1358] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[1359] Step 9:

[1360] The application site will be set up so that users can check the results of the screening. Users can check on the website whether their application has been approved or not.

[1361] Example 1

[1362] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1363] Conventional identity verification methods require manual verification of document authenticity, which takes a great deal of time and effort. It also makes it difficult to detect fraudulent applications early, resulting in low accuracy and efficiency of screening. Furthermore, with the spread of online applications, there is a growing need for digitization and automated processing of documents, but current systems are inadequate to meet this demand.

[1364] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1365] In this invention, the server includes: means for extracting character data from identification documents using OCR technology; means for analyzing the characteristics of the extracted character data using an AI algorithm; means for comparing the analyzed character data characteristics with an existing database to determine authenticity; means for notifying the review results based on the results of the review; means for the user to upload an image of the identification document to the online application system using a terminal; means for the server to store and preprocess the received image data; means for providing the preprocessed image to an OCR engine to extract text data; means for analyzing the character arrangement, font, position, size, etc. using an AI algorithm; means for calculating the degree of match based on the analysis results and classifying the result as "true," "false," or "suspicious"; and means for notifying a manual review team if the result is suspicious. This enables automatic and highly accurate determination of the authenticity of identification documents and early detection of fraudulent applications. It also improves the efficiency and accuracy of review and can be easily incorporated into online application systems.

[1366] "OCR technology" is optical character recognition technology, which extracts character data from images.

[1367] "AI algorithm" refers to a calculation method or model that uses artificial intelligence, and is a technology used to analyze and process data.

[1368] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[1369] "Image data" refers to image information stored in digital format, and includes images of identification documents.

[1370] "Preprocessing" refers to processing performed to analyze image data in optimal conditions, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[1371] "Text data" refers to character information extracted from image data and is obtained using OCR technology.

[1372] "Feature analysis" is the process of analyzing various aspects or characteristics of data, including character alignment, font, position, size, etc.

[1373] "Database" refers to a system or collection for systematically managing and storing large amounts of data, in which authentic identity verification data is stored.

[1374] "Concordance" is an index that indicates how closely the analysis results match the records in the database.

[1375] A "judgment result" is a conclusion based on analysis and degree of agreement, and is classified as "true," "false," or "doubtful."

[1376] The "Manual Review Team" is a specialized team that performs detailed human review of suspicious judgment results.

[1377] "Online application system" refers to a system that allows various applications and procedures to be carried out via the Internet.

[1378] A "server" is a computer system that stores, processes, and provides data over a network.

[1379] "Terminal" refers to a device such as a computer or smartphone operated by a user.

[1380] This invention relates to a system that automatically determines the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract character data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It also includes a function that notifies the user of the results of the examination based on the results.

[1381] System configuration

[1382] The main components of the system are:

[1383] 1. User device (PC or smartphone)

[1384] 2. Server

[1385] 3. Online Application System

[1386] 4. OCR engine (e.g. Tesseract OCR)

[1387] 5. AI Algorithms

[1388] 6. Database

[1389] Operation overview

[1390] A user uses a terminal to upload an image of their identification document, such as a driver's license, through an online application system. The server then stores the received image data. The stored image data is preprocessed to resize, remove noise, straighten, and adjust brightness and contrast. This preprocessing is necessary to improve the accuracy of OCR analysis.

[1391] After preprocessing, the image is analyzed by an OCR engine to extract text data. The server saves this text data as a text file. The extracted text data is then input into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[1392] Judgment and Notification

[1393] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, it is notified to the manual review team, who then conduct a visual review. Finally, the result is notified to the application system, and the results are shared with the user.

[1394] Specific operation example

[1395] User operation

[1396] A user applies for a new mobile phone contract online.

[1397] The user uploads an image of their driver's license and enters the required personal information.

[1398] Server processing

[1399] The server receives image data from the user and stores it in a database.

[1400] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[1401] The preprocessed image is fed to an OCR engine to extract text data.

[1402] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[1403] Judgment and Notification

[1404] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[1405] The results are classified as "true," "false," or "suspicious." If the result is suspicious, an alert is sent to the eye inspection team.

[1406] The server notifies the application system of the judgment result and displays the examination result to the user.

[1407] As a concrete example, the following prompt sentence is input to a generative AI model:

[1408] Example prompt sentence:

[1409] Please explain in detail the processing steps of the system for determining the authenticity of ID documents. The user uploads an image of their driver's license, the server performs preprocessing and OCR analysis, and then an AI algorithm analyzes the features to determine authenticity.

[1410] This system automatically determines the authenticity of identification documents, significantly improving the accuracy and efficiency of screening. It also enables early detection of fraudulent applications, reducing financial risks for companies. The system based on this invention can be easily incorporated into existing online application systems and is applicable to a wide range of uses.

[1411] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1412] Step 1:

[1413] User upload of identification documents

[1414] A user uses a terminal (PC or smartphone) to access the online application system and uploads an image of an identification document (e.g., a driver's license).

[1415] Input: Image of user's ID document

[1416] Output: Image data uploaded to the online application system

[1417] Step 2:

[1418] Receiving and storing image data by the server

[1419] The server receives the image data uploaded from the online application system. The received image data is stored on the server, and metadata linked to the user ID and application details is also recorded in the image data.

[1420] Input: Uploaded image data and metadata

[1421] Output: Stored image data and associated metadata

[1422] Step 3:

[1423] Preprocessing of image data by the server

[1424] The server performs pre-processing on the stored image data, which includes the following specific operations:

[1425] Resize: Unify image resolution to improve processing efficiency.

[1426] Noise Reduction: Remove unwanted noise in the image.

[1427] Deskew: Corrects the tilt of the image so that the text is level.

[1428] Brightness and Contrast Adjustment: Properly adjust the brightness and contrast of the image to improve the accuracy of OCR analysis.

[1429] Input: Saved image data

[1430] Output: Preprocessed image data

[1431] Step 4:

[1432] Server-based OCR analysis and text data extraction

[1433] After preprocessing, the image data is input to an OCR engine (e.g., Tesseract OCR) to extract text data from the image. The server saves this text data as a text file.

[1434] Input: Preprocessed image data

[1435] Output: Extracted text data

[1436] Step 5:

[1437] Feature analysis using AI algorithms on the server

[1438] The server inputs the extracted text data into an AI algorithm to analyze features such as character arrangement, font, position, and size. The analysis results provide detailed feature information about the text data.

[1439] Input: Extracted text data

[1440] Output: Analyzed feature information

[1441] Step 6:

[1442] Comparison with database by server and calculation of match

[1443] The server compares the analysis results with an existing database and calculates the degree of match, which determines how closely the text matches the authentic text.

[1444] Input: Analyzed feature information and normal data in the database

[1445] Output: Match rating

[1446] Step 7:

[1447] Classification and notification of judgment results by the server

[1448] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is "suspicious," it sends an alert to the manual review team. Finally, the result is notified to the online application system, which then shares the results with the user.

[1449] Input: Match evaluation result

[1450] Output: Classified judgment results and notification information

[1451] For example, when a user applies for a new mobile phone contract, they upload an image of their driver's license. The server receives the image, performs preprocessing and OCR analysis, and then performs AI feature analysis and comparison with a database. If the AI ​​algorithm determines the image to be "suspicious," an alert is sent to the manual review team. After this process, the user is finally notified of the result.

[1452] (Application example 1)

[1453] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1454] Conventional identity verification systems require manual verification, which requires a great deal of time and resources. They also lack reliable methods for preventing the use of fraudulent identity documents. While electronic payment services require the immediate registration of new accounts, they also require highly accurate verification of the authenticity of identity documents. Therefore, there is a growing need for a system that can efficiently and automatically verify the authenticity of identity documents and immediately notify users of the results.

[1455] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1456] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for determining authenticity by comparing the characteristics of the analyzed character data with an existing database, means for notifying the result of the examination based on the determination result, means for transmitting image data of the identification document from the mobile device to the server at the time of new registration, and means for extracting text data from the preprocessed image data and analyzing the characteristics of the character string using a model. This makes it possible to efficiently and automatically determine the authenticity of identification documents in electronic payment services and immediately notify the user of the result.

[1457] "OCR technology" stands for optical character recognition technology, which is a technology that automatically extracts characters from images.

[1458] An "AI algorithm" is an algorithm that uses artificial intelligence, and is a method of automatically learning and making decisions based on large amounts of data.

[1459] "Identification document" refers to an official document used to verify an individual's identity, such as a driver's license or passport.

[1460] "Text data" refers to text information extracted from an image using OCR technology.

[1461] "Feature analysis" is the process of analyzing the characteristics of extracted character data, such as the arrangement, font, position, and size.

[1462] The "existing database" is a database that stores previously accumulated characteristic data of identification documents.

[1463] "Authenticity determination" is the process of determining whether a document is genuine or fake by comparing the characteristics of the analyzed character data with existing databases.

[1464] "Notification of inspection results" is the process of notifying the user or system of the results of the authenticity determination.

[1465] A "mobile device" is a portable electronic device such as a smartphone or tablet.

[1466] A "server" is a computer system that processes data, analyzes image and text data, and manages the results of judgments.

[1467] "Preprocessing" is the process of resizing, removing noise, correcting tilt, and adjusting brightness and contrast to make image data easier to analyze.

[1468] "Model" means a machine learning model trained by an AI algorithm and used to analyze data.

[1469] This invention is a system that efficiently and automatically determines the authenticity of identity documents in electronic payment services and immediately notifies the user of the results. This system eliminates the need for the conventional manual verification process and enables rapid and accurate identity verification.

[1470] System configuration and operation overview

[1471] 1. User Action:

[1472] When a user creates a new account for an electronic payment service, the user takes a picture of their identification document (e.g., driver's license) using a mobile device such as a smartphone and sends it to a server via an application.

[1473] 2. Server Processing:

[1474] The server receives the image data of the ID document sent by the user and performs preprocessing, including image resizing, noise removal, tilt correction, brightness and contrast adjustment, etc. This is done using OpenCV, an open source image processing library.

[1475] After preprocessing, the image is then processed using the OCR engine Pytesseract to extract text data, which contains the necessary character information and is then analyzed using AI algorithms.

[1476] The extracted character data is fed into a pre-trained machine learning model, which is trained using a machine learning library such as scikit-learn, to analyze specific features of the string (e.g., character alignment, font, position, size, etc.).

[1477] 3. Authentication and Notification:

[1478] The server compares the characteristics of the analyzed text data with an existing database and calculates the degree of match. Based on this degree of match, the authenticity is classified into three categories: "true," "false," or "suspicious." If the authenticity is doubtful, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified.

[1479] The result of the check is immediately notified to the user via the application, for example, a message such as "Your identity verification documents are valid, so your account creation has been completed."

[1480] Specific use cases

[1481] Register a new account:

[1482] A user launches an application to create a new electronic payment account and uploads a photo of their driver's license.

[1483] The server receives the images, pre-processes them, and analyzes them using OCR and AI algorithms.

[1484] The result of the judgment is immediately notified to the user, and if it is true, the account creation is completed.

[1485] Prompt Sentence Examples

[1486] To create a new e-payment account, please upload a photo of your driver's license. The system will automatically verify the authenticity of your ID and notify you of the processing result.

[1487] The system is expected to speed up the process of registering new accounts for electronic payment services, improve the accuracy of identity verification, and significantly improve the user experience.

[1488] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1489] Step 1:

[1490] To create a new electronic payment account, a user takes a photo of their identity document with their mobile device and uploads the image to a server via an application. The input is the image data of the identity document, which is then sent to the server. The output is the image data stored on the server.

[1491] Step 2:

[1492] The server performs preprocessing on the received image data. Specifically, it uses OpenCV to resize, remove noise, correct distortion, and adjust brightness and contrast. The input is the uploaded image data, and the output is the preprocessed image data.

[1493] Step 3:

[1494] The server supplies the preprocessed image data to an OCR engine (Pytesseract) to extract character data. The input is the preprocessed image data, and the output is the extracted character data. At this stage, character information is obtained in text format.

[1495] Step 4:

[1496] The extracted character data is input into an AI algorithm for feature analysis. Specifically, a machine learning library (such as scikit-learn) is used to analyze features such as character arrangement, font, position, and size. The input is text data, and the output is the analyzed feature data.

[1497] Step 5:

[1498] The server compares the analyzed data with the existing database and calculates the degree of match. The input is the analyzed data and the existing database, and the output is a score of the degree of match. In this step, an authentication algorithm based on a specific model is used.

[1499] Step 6:

[1500] Based on the degree of match, the authenticity is judged and the judgment result is classified into three categories: "true," "false," or "suspicious." The input is the match score, and the output is the status of the judgment result. In this step, if the result is suspicious, a flag is set indicating that manual review is required.

[1501] Step 7:

[1502] The server notifies the user of the final review result through the application system. Specifically, the review result is displayed on the user's mobile device. The input is the review result status, and the output is the review result displayed to the user. This result allows the user to confirm whether or not the account creation is complete.

[1503] The above steps effectively automate the new account registration process for electronic payment services.

[1504] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1505] This invention combines an emotion engine with a system for automatically determining the authenticity of identification documents. This system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of that data, and compare it with an existing database to determine authenticity. It then combines an emotion engine that recognizes the user's emotional state and optimizes the notification process of the screening results based on the emotional information.

[1506] System configuration and operation overview

[1507] First, a user uploads an image of their identification document, such as a driver's license, through an online application system. This operation is performed on the user's device. The server then stores the received image data and performs preprocessing, which includes image resizing, noise reduction, tilt correction, and brightness and contrast adjustment.

[1508] After the server performs these preprocessing steps, the OCR engine extracts text data from the image. The extracted text data is saved as a text file on the server. This text data is then input into an AI algorithm, which analyzes features such as character arrangement, font, position, and size. The analysis results are compared with an existing database to calculate the degree of match.

[1509] Based on this degree of match, the server classifies the judgment result as "true," "false," or "suspicious." If the result is suspicious, a flag is set indicating that a visual inspection is required, and the manual inspection team is notified. The judgment result is also notified to the application system, and the result is shared with the user. At the same time, the emotion engine analyzes the user's emotional state and reflects this information in the inspection process.

[1510] Overview of Emotion Engine Operation

[1511] The emotion engine analyzes the user's emotional state in real time based on camera images, audio data, timing information of input operations, etc. The results of this analysis are sent to the server and used to notify the review results and for follow-up procedures.

[1512] Specific example of system operation

[1513] 1. User Actions

[1514] A user applies for a new mobile phone contract online.

[1515] The user uploads an image of their driver's license and enters the required personal information.

[1516] 2. Server Processing

[1517] The server receives image data from the user and stores it in a database.

[1518] The server starts pre-processing the image, resizing, denoising, deskewing, and adjusting brightness and contrast.

[1519] The preprocessed image is fed to an OCR engine to extract text data.

[1520] The extracted text data is input into an AI algorithm to analyze the characteristics of the characters.

[1521] The server compares the AI ​​analysis results with the database and calculates the degree of match.

[1522] 3. Coordination of judgment and emotion analysis

[1523] The server classifies the results as "true," "false," or "suspicious" based on the degree of match.

[1524] The server starts the emotion engine and collects and analyzes the user's emotion data.

[1525] The server integrates the analysis results and sends the review results to the application system. Emotional data is used to customize notification methods and message content.

[1526] 4. Notice to Users

[1527] The user checks the results of the review on the application site, and the notification includes appropriate feedback that reflects the results of the sentiment analysis.

[1528] If necessary, follow-up suggestions are provided to the user.

[1529] This system based on the present invention not only improves the accuracy of determining the authenticity of identification documents, but also improves the user experience and further prevents fraud by enabling flexible notification of review results that take into account the user's emotional state.

[1530] The processing flow will be explained below.

[1531] Step 1:

[1532] The user uploads an image of their identification (e.g., driver's license) through an online application site. The user enters the required personal information and submits the application form.

[1533] Step 2:

[1534] The server receives the image data sent by the user and stores the received image data in a database.

[1535] Step 3:

[1536] The server performs preprocessing on the stored image data. The preprocessing includes the following operations:

[1537] Image resizing: Adjust the size so that OCR and AI can process it properly.

[1538] Denoise: Apply filtering to reduce noise in the image.

[1539] Deskew: If the image is tilted, correct the tilt.

[1540] Brightness and contrast adjustment: Adjust brightness and contrast to improve character recognition accuracy.

[1541] Step 4:

[1542] The server inputs the preprocessed image data into the OCR engine, which extracts text data from the image. This data includes the name, address, date of birth, etc. The server saves the extracted text data as a text file for further processing.

[1543] Step 5:

[1544] The server inputs the text data extracted by OCR into an AI algorithm, which analyzes character features (arrangement, font, position, size, etc.). Specifically, it evaluates the balance of character spacing and position, font type and size, and consistency of character strings.

[1545] Step 6:

[1546] The server compares the analysis results with a database of existing identity documents. The AI ​​compares them with existing data and calculates the degree of match of character features. If a mismatch is detected, the details are also recorded.

[1547] Step 7:

[1548] The server performs the following classification based on the comparison results:

[1549] "True": If there is a high degree of match with existing data.

[1550] "False": If the match is very low.

[1551] "Doubtful": intermediate agreement or significant discrepancies in specific areas.

[1552] Step 8:

[1553] The server sends the result of the judgment to the application system. If the result is "true," the application system is given permission to proceed to the next screening step. If the result is "false" or "suspicious," the application is rejected or notified that it will be subject to visual inspection.

[1554] Step 9:

[1555] The server starts the emotion engine and collects camera images, audio data, and timing information of input operations from the user's device. The emotion engine uses this data to analyze the user's emotional state in real time.

[1556] Step 10:

[1557] The server integrates the emotion analysis results into the review result notification process, and optimizes the content and method of review result notification taking into account the user's emotional state.

[1558] Step 11:

[1559] The server sends the final review results, feedback based on the emotional state, and follow-up suggestions, if necessary, to the application system.

[1560] Step 12:

[1561] The user checks the results of the review on the application site. The notification includes appropriate feedback that reflects the results of the sentiment analysis. The user can then check the results and take the next action.

[1562] Example 2

[1563] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1564] In conventional identity verification systems, determining the authenticity of identity documents is often done manually, which is time-consuming and labor-intensive. Even in automated systems, there is no notification process that takes into account the user's emotional state, making it difficult to improve the user experience. Furthermore, the lack of emotion analysis means that notification of results is uniform, making it difficult to respond flexibly.

[1565] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1566] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity, emotion engine means for analyzing the user's emotional state in real time, means for optimizing the notification process of the examination result based on the analysis result of the emotion engine means, and means for notifying the examination result based on the determination result. This improves the accuracy of authenticity determination and enables a flexible notification process based on the user's emotional state, significantly improving the user experience.

[1567] "OCR technology" stands for optical character recognition technology, which recognizes characters from image data and extracts them as text data.

[1568] "Identification document" means an official document used to prove an individual's identity, such as a driver's license or passport.

[1569] An "AI algorithm" is an algorithm based on artificial intelligence technology that learns, analyzes, and predicts specific tasks.

[1570] An "emotion engine" is a system for analyzing a user's emotional state, and is a technology that recognizes emotions in real time based on input information such as camera images and audio data.

[1571] "Preprocessing" refers to processing performed to prepare image data in a format that is easier to analyze, and includes resizing, noise removal, tilt correction, brightness and contrast adjustment, etc.

[1572] "Resizing" is an operation for changing the resolution and size of image data, and is a process for adjusting the resolution to a specified value.

[1573] "Noise reduction" is a process that removes unnecessary information and dust from image data, and is performed to improve image quality.

[1574] "Tilt correction" is a process of correcting the tilt of image data and leveling it.

[1575] "Brightness and contrast adjustment" is a process for adjusting the brightness and contrast of image data to improve visibility.

[1576] The "inspection result" is a result including a determination of the authenticity of the personal identification document and notification content based on the user's emotional state.

[1577] "Authenticity determination" is the process of determining whether an identification document is genuine or fake.

[1578] This invention combines a system for automatically determining the authenticity of identification documents with an emotion engine that analyzes the user's emotional state. The system uses OCR technology and AI algorithms to extract text data from identification documents, analyze the characteristics of the data, and compare it with existing databases to determine authenticity. It then recognizes the user's emotional state in real time and optimizes the notification process of the screening results based on the emotional information.

[1579] System configuration and operation overview

[1580] 1. User Actions

[1581] The user accesses the online application system using an internet-connected device (e.g., smartphone, PC) and takes a photo of their identification document (e.g., driver's license, passport) or selects and uploads an existing image file. This action begins the first step in the identity verification process.

[1582] 2. Server Processing

[1583] The server stores the image data received from the user and then pre-processes it, which includes the following steps:

[1584] Resize: Change the image resolution to the specified size.

[1585] Denoising: Removing unnecessary information and dust from an image, for example using a Gaussian filter.

[1586] Deskew: Level out the tilt of the image, for example by using the Hough transform.

[1587] Brightness and Contrast Adjustment: Adjust the image brightness and contrast to the optimum level.

[1588] 3. Use of OCR technology

[1589] The preprocessed image is fed to an OCR engine (e.g., Tesseract OCR) to extract text data, which is then saved to a server as a text file.

[1590] 4. Analysis by AI algorithm

[1591] The extracted text data is input into an AI algorithm (e.g., TensorFlow model) to analyze features such as character arrangement, font, position, size, etc. The analysis results are compared with an existing database on the server to calculate the degree of match.

[1592] 5. Authenticity Verification

[1593] The server classifies the results as "true," "false," or "suspicious" based on the degree of match. If the result is suspicious, a flag is set indicating that a visual review is required, and the manual review team is notified.

[1594] 6. Emotion Analysis Using an Emotion Engine

[1595] The server analyzes the user's emotional state in real time based on camera images, audio data, and timing information of input operations acquired from the user's device. This analysis is performed using an emotion engine (e.g., Affectiva SDK). The results of this analysis are sent to the server and used to notify the user of the review results and for follow-up procedures.

[1596] 7. Notification of the review results

[1597] The server integrates the analysis results with the user's emotional state to generate an optimized notification message. The review results are notified to the user through the application system. The notification includes appropriate feedback that reflects the user's emotional state.

[1598] Specific example of system operation

[1599] When a user applies online for a new mobile phone contract, they upload a picture of their driver's license.

[1600] The server receives the image and stores it in a database.

[1601] Preprocessing involves resizing the image, removing noise, straightening the image, and adjusting the brightness and contrast.

[1602] An OCR engine is used to extract text data, which is then analyzed using AI algorithms.

[1603] The server compares the analysis results with a database and calculates the degree of match.

[1604] Based on the degree of match, the authenticity of the document is determined, and an emotion engine is used to collect and analyze user emotion data.

[1605] Finally, the server integrates the analysis results with the emotional information and sends the results of the screening to the application system as an optimized notification message. The user can then check the results on the application site and receive follow-up instructions if necessary.

[1606] Prompt Sentence Examples

[1607] Here are some example prompts to input to a generative AI model:

[1608] When a user uploads a photo of their driver's license, the system automatically pre-processes the data and uses OCR technology and AI algorithms to extract and analyze text data to determine the authenticity of the ID document. It also analyzes the user's emotional state to optimize the notification process of the review results.

[1609] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1610] Processing Step Description

[1611] Step 1:

[1612] Input: The user uses the device to upload an image file of their identification document (e.g., driver's license).

[1613] Specific operation: A user accesses the online application system, selects an image, and clicks the upload button.

[1614] Output: The uploaded image file is sent to the server.

[1615] Step 2:

[1616] Input: The image file received by the server.

[1617] Specific operation: The server receives the image file and stores it in the database.

[1618] Output: Saved image file.

[1619] Step 3:

[1620] Input: Saved image file.

[1621] What it does: The server resizes the image, changing the resolution to the specified size, then applies noise reduction to remove unnecessary information from the image, deskews the image to make it horizontal, and adjusts the brightness and contrast.

[1622] Output: Preprocessed image files.

[1623] Step 4:

[1624] Input: Preprocessed image files.

[1625] Specific operation: The server inputs the preprocessed image into an OCR engine (e.g., Tesseract OCR) and extracts text data from the image.

[1626] Output: The extracted text data.

[1627] Step 5:

[1628] Input: Extracted text data.

[1629] How it works: The server inputs text data into an AI algorithm (e.g., a TensorFlow model) and analyzes features such as character alignment, font, position, and size.

[1630] Output: Features of the parsed character data.

[1631] Step 6:

[1632] Input: Parsed character data features.

[1633] How it works: The server compares the analysis results with the existing database and calculates the degree of match. In this process, it calculates how closely the character arrangement and font match the existing data.

[1634] Output: Match data.

[1635] Step 7:

[1636] Input: Match data.

[1637] How it works: The server classifies the result as "true," "false," or "suspicious" based on the degree of match. If it is suspicious, it sets a flag indicating that manual review is required.

[1638] Output: Classified judgment result.

[1639] Step 8:

[1640] Input: Camera images and audio data obtained from the user's device, as well as timing information for input operations.

[1641] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's emotional state in real time.

[1642] Output: User's emotional state data.

[1643] Step 9:

[1644] Input: Judgment result and user's emotional state data.

[1645] Specific operation: The server integrates the judgment result and emotional information to generate an optimized notification message.

[1646] Output: Informational message.

[1647] Step 10:

[1648] Input: Notification message.

[1649] Specific operation: The server sends a notification message reflecting the review result and emotional state to the user through the application system.

[1650] Output: The final review result for the user to review.

[1651] Through these steps, the server can determine the authenticity of the identification documents and notify the user of the results of the review, taking into account the user's emotional state, thereby improving the user experience.

[1652] (Application example 2)

[1653] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1654] Food delivery services are required to quickly and accurately determine the authenticity of delivery partners' identification documents. It is also important to improve the quality of service by taking into account the emotional state of delivery personnel and users and responding appropriately. Current systems do not have the functionality to fully resolve these issues, so a mechanism is needed to efficiently increase safety and user satisfaction.

[1655] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1656] In this invention, the server includes means for extracting character data from identification documents using OCR technology, means for analyzing the characteristics of the extracted character data using an AI algorithm, and means for analyzing the emotional state of the user using emotion analysis technology, which enables the authenticity of the delivery partner's identification documents to be determined and optimal notifications to be given based on the emotional state of the user or delivery person.

[1657] "OCR technology" is an optical character recognition technology that analyzes character information input as an image and reads it as digital text.

[1658] An "AI algorithm" is a type of artificial intelligence technology, and refers to a calculation procedure for analyzing data and performing pattern recognition and predictions.

[1659] "Emotion analysis technology" is a technology that analyzes and judges a user's emotional state based on camera images and audio data.

[1660] "Identification documents" are documents used to prove an individual's identity or qualifications, such as a driver's license or passport.

[1661] "Authentication" refers to the process of determining whether an item or data is genuine or counterfeit.

[1662] A "server" is a computer system that serves the purpose of providing services to other computers on a network.

[1663] "Preprocessing" refers to a series of processing steps performed on raw data, including data cleansing and transformation before analysis or recognition.

[1664] "Resizing" refers to the operation of changing the size of an image or data.

[1665] "Noise reduction" is a technology that removes unnecessary information from image and audio data to improve the quality of the data.

[1666] "Tilt correction" is the process of straightening out images or text that are tilted.

[1667] "Brightness and contrast adjustment" refers to the adjustment process used in image processing to optimize the brightness and contrast of an image.

[1668] A "database" is a system for efficiently managing, searching, and updating large amounts of data.

[1669] "Notification" is the process of conveying information or results to a specific audience.

[1670] This invention is a system that verifies the authenticity of delivery partners' identification documents in food delivery services and analyzes the emotional state of users and delivery partners. This system combines OCR technology, AI algorithms, and emotion analysis technology to provide a safe service that provides high levels of user satisfaction.

[1671] System configuration

[1672] The system consists of devices used by users and delivery partners, a server that performs image processing and data analysis, and a cloud service for emotion analysis.

[1673] Hardware and Software

[1674] Device: A smartphone or tablet that allows users and delivery partners to upload images of their identification documents.

[1675] Server: A computer system that stores data, performs preprocessing, performs AI analysis, and notifies results. For example, cloud services (AWS, Google Cloud Platform) are used.

[1676] OCR technology: Extracts text data from images using Google Cloud Vision API.

[1677] AI algorithm: TensorFlow is used to analyze the characteristics of the extracted text data and determine its authenticity.

[1678] Emotion analysis technology: Uses Microsoft Azure Face API to analyze the emotional state of users and delivery partners.

[1679] Data processing and calculation

[1680] The server receives image data of identification documents uploaded by users and delivery partners, and performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment. Next, text data is extracted from the preprocessed image data using OCR technology using the Google Cloud Vision API. The extracted text data undergoes feature analysis using TensorFlow and is compared with an existing database to determine authenticity. Sentiment analysis technology is also used to analyze the emotional state of users and delivery partners in real time.

[1681] Specific examples

[1682] For example, when a delivery partner registers, they upload an image of their driver's license. This image is sent to a server, where it is pre-processed and text data is extracted using OCR technology. It is then analyzed by an AI algorithm to determine its authenticity. At the same time, the delivery partner's emotional state is analyzed from the camera image, and an appropriate notification message is sent based on that emotional state.

[1683] Prompt Sentence Examples

[1684] "Please upload an image of the delivery partner's verification document. We use OCR technology to extract the text data, and AI will determine its authenticity. Next, we will perform sentiment analysis using the camera image and provide the most appropriate notification."

[1685] In this way, the system combines the authentication of identification documents with the analysis of emotional states, making the delivery partner registration process smooth and safe, while also enabling notifications that take into account the emotional state of users and delivery partners.

[1686] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1687] Step 1:

[1688] The user uploads an image of their identity document from their device.

[1689] Input: Image of identification document

[1690] Specific operation: The user uses a smartphone or tablet to take a photo of their identification document, such as a driver's license or passport, and uploads the image using a dedicated application.

[1691] Step 2:

[1692] The terminal transmits the uploaded image data to the server.

[1693] Input: Image of identification document

[1694] Output: Image data transferred to the server

[1695] Specific operation: The device sends the uploaded image data to the server, which receives the data and stores it in a database.

[1696] Step 3:

[1697] The server performs pre-processing of the image data.

[1698] Input: Uploaded image data

[1699] Output: Preprocessed image data

[1700] Specific operation: The server performs preprocessing such as resizing, noise removal, tilt correction, and brightness and contrast adjustment, thereby obtaining image data suitable for analysis.

[1701] Step 4:

[1702] The server extracts character data from the preprocessed image data using OCR technology.

[1703] Input: Preprocessed image data

[1704] Output: Extracted character data

[1705] How it works: The server uses the Google Cloud Vision API to extract text information from the preprocessed image data, and the extracted text data is stored on the server.

[1706] Step 5:

[1707] The server uses an AI algorithm to analyze the extracted text data.

[1708] Input: Extracted character data

[1709] Output: Analysis results (characteristics of character data)

[1710] Specific operation: The server uses TensorFlow to analyze the extracted character data's characteristics, such as alignment, font, position, and size. The analysis results are stored in a database on the server.

[1711] Step 6:

[1712] The server compares the analysis results with an existing database to determine authenticity.

[1713] Input: Analysis results (characteristics of character data)

[1714] Output: Matching result (judgment result)

[1715] Specific operation: The server compares the analysis results with the existing database and calculates the degree of match. Based on the calculated degree of match, it classifies the result as "true," "false," or "suspicious."

[1716] Step 7:

[1717] The server analyzes the user's emotions using emotion analysis technology.

[1718] Input: User's camera image

[1719] Output: Emotion analysis results

[1720] How it works: The server uses the Microsoft Azure Face API to analyze the user's emotional state from their camera image. The analyzed emotional state is then stored on the server.

[1721] Step 8:

[1722] The server notifies the review results based on the judgment results and emotion analysis results.

[1723] Input: Matching result (judgment result), emotion analysis result

[1724] Output: Notification message

[1725] Specific operation: The server combines the matching results and the sentiment analysis results to generate and send the optimal notification message to the user. For example, it customizes the tone and content of the notification message based on the sentiment analysis results.

[1726] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1727] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1728] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1729] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1730] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1731] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1732] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1733] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1734] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1735] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1736] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1737] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1738] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1739] 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.

[1740] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1741] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1742] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1743] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1744] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1745] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1746] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1747] The following is further disclosed regarding the above embodiment.

[1748] (Claim 1)

[1749] A means for extracting character data from an identification document using OCR technology;

[1750] A means for analyzing the characteristics of extracted text data using an AI algorithm;

[1751] A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity;

[1752] A means for notifying the results of the examination based on the judgment results;

[1753] A system including:

[1754] (Claim 2)

[1755] 2. The system according to claim 1, further comprising a pre-processing means for resizing, denoising, skew correcting, and adjusting brightness and contrast of the image data of the identification document.

[1756] (Claim 3)

[1757] 2. The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis result of the character data.

[1758] "Example 1"

[1759] (Claim 1)

[1760] A means for extracting character data from an identification document using OCR technology;

[1761] A means for analyzing the characteristics of extracted text data using an AI algorithm;

[1762] A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity;

[1763] A means for notifying the results of the examination based on the judgment results;

[1764] A means for a user to upload an image of an identification document to an online application system using a terminal;

[1765] means for storing and preprocessing the image data received by the server;

[1766] A means for supplying the preprocessed image to an OCR engine and extracting text data;

[1767] A method to analyze character arrangement, font, position, size, etc. using AI algorithms,

[1768] A means for calculating the degree of coincidence based on the analysis results and classifying the judgment results into "true," "false," or "suspicious";

[1769] A system that includes a means to notify a manual review team in cases of suspicion.

[1770] (Claim 2)

[1771] 2. The system according to claim 1, further comprising a pre-processing means for resizing, denoising, skew correcting, and adjusting brightness and contrast of the image data of the identification document.

[1772] (Claim 3)

[1773] 2. The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis result of the character data.

[1774] "Application Example 1"

[1775] (Claim 1)

[1776] A means for extracting character data from an identification document using OCR technology;

[1777] A means for analyzing the characteristics of extracted text data using an AI algorithm;

[1778] A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity;

[1779] A means for notifying the results of the examination based on the judgment results;

[1780] A means for transmitting image data of an identification document from a mobile device to a server when registering a new user;

[1781] A means for extracting text data from the preprocessed image data and analyzing the characteristics of the character string using a model;

[1782] A system including:

[1783] (Claim 2)

[1784] The system according to claim 1, further comprising a pre-processing means for resizing, removing noise, correcting skew, and adjusting brightness and contrast of image data of the identification document, and for processing image data uploaded from the mobile device.

[1785] (Claim 3)

[1786] The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis results of the character data, and notifying the mobile device of the classification results.

[1787] "Example 2: Combining Emotion Engines"

[1788] (Claim 1)

[1789] A means for extracting character data from an identification document using OCR technology;

[1790] A means for analyzing the characteristics of extracted text data using an AI algorithm;

[1791] A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity;

[1792] an emotion engine means for analyzing the user's emotional state in real time;

[1793] A means for optimizing the notification process of the review results based on the analysis results of the emotion engine means;

[1794] A means for notifying the results of the examination based on the judgment results;

[1795] A system including:

[1796] (Claim 2)

[1797] 2. The system according to claim 1, further comprising a pre-processing means for resizing, denoising, skew correcting, and adjusting brightness and contrast of the image data of the identification document.

[1798] (Claim 3)

[1799] 2. The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis result of the character data.

[1800] "Application example 2 when combining emotion engines"

[1801] (Claim 1)

[1802] A means for extracting character data from an identification document using OCR technology;

[1803] A means for analyzing the characteristics of extracted text data using an AI algorithm;

[1804] A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity;

[1805] means for analyzing the emotional state of a user using emotion analysis techniques;

[1806] A method for optimizing notification content based on that emotional information, and

[1807] A means for notifying the results of the examination based on the judgment results;

[1808] A system including:

[1809] (Claim 2)

[1810] 2. The system according to claim 1, further comprising a pre-processing means for resizing, denoising, skew correcting, and adjusting brightness and contrast of the image data of the identification document.

[1811] (Claim 3)

[1812] 2. The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis result of the character data. [Explanation of symbols]

[1813] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for extracting character data from an identification document using OCR technology; A means for analyzing the characteristics of extracted text data using an AI algorithm; A means for comparing the characteristics of the analyzed character data with an existing database to determine authenticity; A means for notifying the results of the examination based on the judgment results; A system including:

2. 2. The system according to claim 1, further comprising a pre-processing means for resizing, removing noise, correcting skew, and adjusting brightness and contrast of the image data of the personal identification document.

3. 2. The system according to claim 1, further comprising a classification means for classifying the character data into "true," "false," or "suspicious" based on the analysis result of the character data.

Images