system

Abstract

The system according to this embodiment aims to analyze testimonies, efficiently generate composite images, and improve the speed of investigations. [Solution] The system according to the embodiment comprises a testimony analysis unit, an image generation unit, a fine-tuning unit, a candidate presentation unit, and a database matching unit. The testimony analysis unit analyzes the testimony and extracts the necessary facial features. The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. The fine-tuning unit allows the witness to make fine adjustments based on the image generated by the image generation unit. The candidate presentation unit presents multiple candidates based on the image fine-tuned by the fine-tuning unit. The database matching unit matches the image selected by the witness from the candidates presented by the candidate presentation unit with a criminal database.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there is a problem that manual search data collation is inefficient because it takes time to create montages and eyewitness testimonies are ambiguous.

[0005] The system according to the embodiment aims to analyze testimonies, efficiently generate montage images, and improve the speed of investigation.

Means for Solving the Problems

[0006] The system according to this embodiment comprises a testimony analysis unit, an image generation unit, a fine-tuning unit, a candidate presentation unit, and a database matching unit. The testimony analysis unit analyzes the testimony and extracts the necessary facial features. The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. The fine-tuning unit allows the witness to make fine adjustments based on the image generated by the image generation unit. The candidate presentation unit presents multiple candidates based on the image fine-tuned by the fine-tuning unit. The database matching unit matches the image selected by the witness from the candidates presented by the candidate presentation unit with a criminal database. [Effects of the Invention]

[0007] The system according to this embodiment can analyze testimonies, efficiently generate composite images, and improve the speed of investigations. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9]This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

[0011] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (Tensor Processing Unit).

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0013] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0014] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F manages communication between multiple computers. Examples of communication standards applicable to the communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0016] [First Embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] As shown in FIG. 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. 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. Also, the database 24 and the communication I / F 26 are 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).

[0019] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0023] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0025] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0026] In the smart device 14, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0027] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) The composite sketch creation system according to an embodiment of the present invention is a system that efficiently creates composite sketch images using a generative AI, thereby improving the speed and accuracy of investigations. Conventional composite sketch creation is time-consuming, and the results are uncertain due to the ambiguity of witness testimonies. The present invention solves these problems with the following components. First, natural language processing is used to analyze the witness's testimony and extract the necessary facial features. The generative AI analyzes the testimony and extracts features such as hairstyle, eye shape, and face shape. Next, an image generation AI is used to create a composite sketch image. Based on the extracted features, the AI ​​generates a composite sketch that reflects hairstyle, eye shape, face shape, etc. Furthermore, it has a function that allows the witness to make fine adjustments based on the generated image to improve its completeness. The witness can create a more accurate composite sketch by making fine adjustments to the details while looking at the generated image. Also, if memory is vague, the AI ​​presents multiple candidates, and the witness can select the optimal composite sketch. The created composite sketch image is compared with a criminal database. The AI ​​compares it with images in the database and identifies individuals who may match. Furthermore, this system can also be used as an aid in character design, allowing designers and creators to gain ideas for character creation. This system improves the speed of investigations and makes the most of witness memories. Additionally, its ease of use in generating character design ideas has led to increased demand in the design market. Thus, the montage creation system can improve the speed and accuracy of investigations.

[0029] The montage creation system according to this embodiment comprises a testimony analysis unit, an image generation unit, a fine-tuning unit, a candidate presentation unit, and a database matching unit. The testimony analysis unit analyzes the testimony and extracts necessary facial features. The testimony analysis unit analyzes the testimony using, for example, a generation AI and extracts features such as hairstyle, eye shape, and face shape. The generation AI can analyze the testimony using natural language processing technology and extract necessary facial features. The image generation unit generates a montage image based on the features extracted by the testimony analysis unit. The image generation unit generates a montage that reflects hairstyle, eye shape, face shape, etc., using, for example, a generation AI. The generation AI can generate a montage image based on the extracted features using image generation technology. The fine-tuning unit allows the witness to make fine adjustments based on the image generated by the image generation unit. The fine-tuning unit allows the witness to make fine adjustments to the details while looking at the generated image, for example, to create a more accurate montage. The fine-tuning unit can also make fine adjustments using AI. The candidate presentation unit presents multiple candidates based on images that have been fine-tuned by the fine-tuning unit. For example, if memory is vague, the candidate presentation unit can present multiple candidates, allowing the witness to select the best one to create the optimal composite image. The candidate presentation unit can also use AI to present candidates. The database matching unit matches the image selected by the witness from the candidates presented by the candidate presentation unit against a criminal database. For example, the database matching unit can use AI to match images in the database and identify individuals who may match. As a result, the composite image creation system according to this embodiment can improve the speed and accuracy of investigations by analyzing testimony, generating composite images, allowing witnesses to make fine adjustments, presenting multiple candidates, and matching them against a database.

[0030] The testimony analysis unit analyzes testimonies and extracts necessary facial features. For example, the testimony analysis unit uses generative AI to analyze testimonies and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze testimonies using natural language processing technology and extract necessary facial features. Specifically, the testimony analysis unit inputs testimonies obtained from eyewitnesses as text data, and the generative AI analyzes this text data. The generative AI extracts keywords and phrases related to facial features from the testimonies and identifies facial features based on them. For example, it analyzes descriptions such as "long hair," "round eyes," and "thin nose," quantifies each feature, and stores it in a database. Furthermore, the generative AI understands the context of the testimonies and can integrate multiple testimonies in a way that does not contradict each other. As a result, the testimony analysis unit can efficiently and accurately analyze eyewitness testimonies and extract necessary facial features. The testimony analysis unit also has a function to evaluate the reliability of testimonies and calculates a reliability score based on the consistency and detail of the testimonies. As a result, the testimony analysis unit can prioritize the analysis of highly reliable testimonies and extract more accurate facial features.

[0031] The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. For example, the image generation unit uses a generation AI to generate a composite image that reflects features such as hairstyle, eye shape, and face shape. The generation AI can generate a composite image based on extracted features using image generation technology. Specifically, the image generation unit inputs facial feature data provided by the testimony analysis unit, and the generation AI combines the various facial parts based on that data to generate a composite image. The generation AI selects parts such as hairstyle, eye shape, nose shape, and mouth shape based on a vast database of pre-trained facial images to generate a natural-looking face image. Furthermore, the generation AI adjusts the placement and size of the facial parts to create an accurate composite image based on the witness's testimony. The image generation unit displays the generated composite image in real time for the witness to review. The image generation unit also saves the generated composite image for use in subsequent processing. This allows the image generation unit to quickly and accurately generate a composite image based on the feature data provided by the testimony analysis unit.

[0032] The fine-tuning unit allows the user to make adjustments based on the image generated by the image generation unit. For example, the fine-tuning unit allows the user to create a more accurate montage by making fine adjustments to the details while viewing the generated image. The fine-tuning unit can also perform fine adjustments using AI. Specifically, the fine-tuning unit provides an interface that allows the user to adjust details such as hairstyle, eye shape, nose shape, and mouth shape while viewing the generated montage image. Through the interface, the user can fine-tune the shape and position of each part to create a more accurate montage image. The fine-tuning unit can also automatically make adjustments based on the user's instructions using AI. For example, if the user instructs the AI ​​to "make the eyes a little bigger," the AI ​​will automatically adjust the shape of the eyes based on that instruction. Furthermore, the fine-tuning unit can reflect the user's feedback in real time and display the adjustment results immediately. This allows the fine-tuning unit to create a more accurate montage by allowing the user to make fine adjustments to the details based on the generated montage image.

[0033] The candidate presentation unit presents multiple candidates based on the image fine-tuned by the fine-tuning unit. For example, if the memory is vague, the candidate presentation unit can present multiple candidates, allowing the witness to select the best one to create the optimal montage. The candidate presentation unit can also present candidates using AI. Specifically, the candidate presentation unit generates multiple variations based on the montage image adjusted by the fine-tuning unit and presents them to the witness. The AI ​​generates variations in different angles and facial expressions based on the features of the fine-tuned image, allowing the witness to select. The witness can then select the one that best matches their memory from the presented candidates, creating the optimal montage. The candidate presentation unit can also further narrow down the candidates based on the witness's selection. For example, if the witness selects a specific candidate, the unit further refines the details based on that candidate and presents new candidates. In this way, the candidate presentation unit can assist in creating the optimal montage based on the witness's memory.

[0034] The database matching unit compares the image selected by the witness from the candidates presented by the candidate suggestion unit with the criminal database. The database matching unit can, for example, use AI to compare images in the database and identify potentially matching individuals. Specifically, the database matching unit receives a composite image selected by the witness and compares it with images in the criminal database. Using facial recognition technology, the AI ​​compares the composite image with images in the database to identify potentially matching individuals. The AI ​​analyzes facial features and patterns to perform highly accurate matching. The database matching unit displays the matching results in a list format, sorted in descending order of matching degree. Investigators can use this list to identify suspects and determine the direction of the investigation. Furthermore, the database matching unit updates the matching results in real time, allowing for rapid response when new data is added. This enables the database matching unit to quickly and accurately identify potentially matching individuals based on the composite image selected by the witness and compared with the criminal database.

[0035] The database matching unit can identify individuals who may potentially match by comparing images in the criminal database. For example, the database matching unit can use AI to match images in the database and identify individuals who may potentially match. The database matching unit can also use facial recognition algorithms to match images in the database and identify individuals who may potentially match. For example, the database matching unit can set a similarity threshold and identify individuals with a similarity above that threshold. This allows for the identification of individuals who may potentially match by comparing images in the criminal database.

[0036] The testimony analysis unit can analyze testimonies using generative AI and extract necessary facial features. For example, the testimony analysis unit uses generative AI to analyze testimonies and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze testimonies using natural language processing technology and extract necessary facial features. For example, the generative AI takes text data of testimonies as input and outputs features such as hairstyle, eye shape, and face shape. As a result, by analyzing testimonies with the generative AI, the necessary facial features can be accurately extracted.

[0037] The image generation unit can generate a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI. For example, the image generation unit generates a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI. The generation AI can generate a montage image based on extracted features using image generation technology. For example, the generation AI takes features such as hairstyle, eye shape, and face shape as input and outputs a montage image. In this way, by generating a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI, an accurate montage image can be created.

[0038] The fine-tuning unit can improve the quality of the montage by making adjustments based on the images generated by the witness. For example, the fine-tuning unit can create a more accurate montage by making fine adjustments to the details while the witness is looking at the generated images. The fine-tuning unit can also use AI to make adjustments. For example, the fine-tuning unit can have the witness select a specific part of the image, and the AI ​​will make adjustments to that part. This allows the witness to make adjustments based on the generated images, thereby improving the quality of the montage image.

[0039] The candidate presentation unit presents multiple candidates when memory is unclear, allowing the witness to choose. For example, by presenting multiple candidates when memory is unclear and allowing the witness to select, the optimal composite image can be created. The candidate presentation unit can also use AI to present candidates. For example, the unit generates multiple candidates based on the witness's testimony, and the witness selects from among them. This allows the witness to select the optimal composite image by presenting multiple candidates when memory is unclear.

[0040] The database matching unit can also be used as an aid in character design, allowing designers and creators to gain ideas for character creation. For example, the database matching unit can use AI to assist in character design. To obtain ideas for character creation, the database matching unit can refer to images within the database and provide relevant information. This allows designers and creators to gain ideas for character creation by utilizing the database matching unit as an aid.

[0041] The testimony analysis unit can improve the accuracy of its analysis by referring to the witness's past testimony history during the analysis process. For example, the testimony analysis unit uses a generative AI to refer to the witness's past testimony history and evaluate the reliability of the current testimony. The generative AI can verify the consistency of the testimony and improve its reliability based on the past testimony history. For example, the generative AI evaluates the reliability of the current testimony by referring to the testimony the witness has provided in the past. The generative AI verifies the consistency of the testimony based on the witness's past testimony history. The generative AI improves the reliability of the testimony by analyzing the witness's past testimony history. In this way, the accuracy of testimony analysis can be improved by referring to the witness's past testimony history.

[0042] The testimony analysis unit can apply different analysis algorithms depending on the content of the testimony during the analysis. For example, the testimony analysis unit can use a generative AI to apply different analysis algorithms depending on the content of the testimony. The generative AI can select the optimal analysis algorithm based on the content of the testimony and analyze the testimony. For example, if the eyewitness testimony is detailed, the generative AI will apply a detailed analysis algorithm. If the eyewitness testimony is ambiguous, the generative AI will apply an analysis algorithm suitable for ambiguous testimony. If the eyewitness testimony is contradictory, the generative AI will apply an analysis algorithm to resolve the contradictions. In this way, the accuracy of testimony analysis can be improved by applying different analysis algorithms depending on the content of the testimony.

[0043] The testimony analysis unit can prioritize the analysis of highly relevant testimonies by considering the geographical location information of witnesses during the testimony analysis process. For example, the testimony analysis unit uses a generative AI to consider the geographical location information of witnesses and prioritize the analysis of highly relevant testimonies. The generative AI evaluates the relevance of testimonies based on the geographical location information of witnesses and can prioritize the analysis of highly relevant testimonies. For example, if a witness is close to the crime scene, the generative AI will prioritize the analysis of that testimony. If a witness is far from the crime scene, the generative AI will lower the priority of that testimony. The generative AI evaluates the relevance of testimonies based on the geographical location information of witnesses. In this way, by considering the geographical location information of witnesses, it is possible to prioritize the analysis of highly relevant testimonies.

[0044] The testimony analysis unit can analyze witnesses' social media activity and analyze relevant testimonies during the testimony analysis process. For example, the testimony analysis unit can use generative AI to analyze witnesses' social media activity and analyze relevant testimonies. Based on social media activity, the generative AI can evaluate the reliability of testimonies and analyze relevant testimonies. For example, the generative AI evaluates the reliability of testimonies by analyzing witnesses' social media posts. Based on witnesses' social media activity, the generative AI confirms the consistency of the testimonies. The generative AI evaluates the relevance of testimonies by referring to witnesses' social media activity. In this way, relevant testimonies can be analyzed by analyzing witnesses' social media activity.

[0045] The image generation unit can adjust the level of detail of an image based on the importance of the testimony during image generation. For example, the image generation unit can use a generation AI to adjust the level of detail of an image based on the importance of the testimony. The generation AI can adjust the level of detail of an image based on the importance of the testimony. For example, if the testimony is detailed, the generation AI generates a high-level image. If the testimony is ambiguous, the generation AI generates a low-level image. The generation AI dynamically adjusts the level of detail of the image according to the importance of the testimony. This allows for the generation of more accurate images by adjusting the level of detail of the image based on the importance of the testimony.

[0046] The image generation unit can apply different generation algorithms depending on the category of the testimony during image generation. For example, the image generation unit can use a generation AI to apply different generation algorithms depending on the category of the testimony. The generation AI can select the optimal generation algorithm based on the category of the testimony and generate an image. For example, if the testimony concerns a person, the generation AI applies a person-specific generation algorithm. If the testimony concerns an object, the generation AI applies an object-specific generation algorithm. If the testimony concerns a landscape, the generation AI applies a landscape-specific generation algorithm. In this way, by applying different generation algorithms depending on the category of the testimony, more appropriate images can be generated.

[0047] The image generation unit can determine image priority based on the timing of testimony submission during image generation. For example, the image generation unit uses a generation AI to determine image priority based on the timing of testimony submission. The generation AI can determine image priority based on the timing of testimony submission. For example, if testimony has been recently submitted, the generation AI will prioritize generating images based on that testimony. If the testimony is old, the generation AI will lower its priority. The generation AI dynamically adjusts the image generation priority according to the timing of testimony submission. This allows for the generation of images based on more important testimony by determining image priority based on the timing of testimony submission.

[0048] The image generation unit can adjust the order of images based on the relevance of the testimonies during image generation. For example, the image generation unit uses a generation AI to adjust the order of images based on the relevance of the testimonies. The generation AI can adjust the order of images based on the relevance of the testimonies. For example, if the testimonies are highly relevant, the generation AI will prioritize the generation of images based on those testimonies. If the testimonies are not highly relevant, the generation AI will lower the priority of those testimonies. The generation AI dynamically adjusts the order of image generation according to the relevance of the testimonies. As a result, by adjusting the order of images based on the relevance of the testimonies, it is possible to generate images that are more relevant.

[0049] The fine-tuning unit can select the optimal adjustment method by referring to the witness's past adjustment history during fine-tuning. For example, the fine-tuning unit uses AI to refer to the witness's past adjustment history and select the optimal adjustment method. Based on the past adjustment history, the AI ​​can propose the optimal adjustment method and verify the consistency of the adjustment. For example, the AI ​​proposes the optimal adjustment method based on the adjustments the witness has made in the past. The AI ​​analyzes the witness's past adjustment history and verifies the consistency of the adjustment. The AI ​​improves the accuracy of the adjustment by referring to the witness's past adjustment history. In this way, the optimal adjustment method can be selected by referring to the witness's past adjustment history.

[0050] The fine-tuning unit can customize the means of adjustment based on the witness's current living situation during the fine-tuning process. For example, the fine-tuning unit uses AI to consider the witness's current living situation and customize the means of adjustment. The AI ​​can dynamically customize the means of adjustment based on the witness's living situation. For example, if the witness is busy, the AI ​​provides quick adjustment options. If the witness is relaxed, the AI ​​provides detailed adjustment options. The AI ​​dynamically customizes the means of adjustment according to the witness's living situation. This allows for more appropriate adjustments by customizing the means of adjustment based on the witness's current living situation.

[0051] The fine-tuning unit can select the optimal adjustment method by considering the geographical location information of witnesses during the fine-tuning process. For example, the fine-tuning unit uses AI to consider the geographical location information of witnesses and select the optimal adjustment method. The AI ​​can evaluate the relevance of adjustments based on the geographical location information of witnesses and select the optimal adjustment method. For example, if a witness is close to the crime scene, the AI ​​will prioritize their testimony in the adjustment. If a witness is far from the crime scene, the AI ​​will lower the priority of their testimony. The AI ​​evaluates the relevance of adjustments based on the geographical location information of witnesses. In this way, the optimal adjustment method can be selected by considering the geographical location information of witnesses.

[0052] The fine-tuning unit can analyze the witness's social media activity during fine-tuning and propose adjustment methods. For example, the fine-tuning unit can use AI to analyze the witness's social media activity and propose adjustment methods. Based on the social media activity, the AI ​​can evaluate the reliability of the adjustment and propose adjustment methods. For example, the AI ​​analyzes the witness's social media posts and evaluates the reliability of the adjustment. Based on the witness's social media activity, the AI ​​confirms the consistency of the adjustment. The AI ​​refers to the witness's social media activity and evaluates the relevance of the adjustment. In this way, by analyzing the witness's social media activity, adjustment methods can be proposed.

[0053] The candidate suggestion unit can suggest the most suitable candidate by referring to the witness's past selection history. For example, the candidate suggestion unit can use AI to refer to the witness's past selection history and suggest the most suitable candidate. Based on the past selection history, the AI ​​can propose the most suitable candidate and verify the consistency of the candidates. For example, the AI ​​suggests the most suitable candidate based on the candidates the witness has previously selected. The AI ​​analyzes the witness's past selection history and verifies the consistency of the candidates. By referring to the witness's past selection history, the AI ​​improves the accuracy of the candidates. In this way, the most suitable candidate can be suggested by referring to the witness's past selection history.

[0054] The candidate presentation unit can customize the method of presenting candidates based on the witness's current living situation. For example, the candidate presentation unit uses AI to consider the witness's current living situation and customize the method of presenting candidates. The AI ​​can dynamically customize the method of presenting candidates based on the witness's living situation. For example, if the witness is busy, the AI ​​will provide quick candidate presentations. If the witness is relaxed, the AI ​​will provide detailed candidate presentations. The AI ​​dynamically customizes the means of presenting candidates according to the witness's living situation. This makes it possible to present more appropriate candidates by customizing the method of presenting candidates based on the witness's current living situation.

[0055] The candidate presentation unit can present the most suitable candidate by considering the geographical location information of the witness. For example, the candidate presentation unit can use AI to consider the geographical location information of the witness and present the most suitable candidate. The AI ​​can evaluate the relevance of the candidates based on the geographical location information of the witness and present the most suitable candidate. For example, if the witness is close to the crime scene, the AI ​​will prioritize presenting that witness's testimony as a candidate. If the witness is far from the crime scene, the AI ​​will lower the priority of that witness's testimony. The AI ​​evaluates the relevance of the candidate presentations based on the geographical location information of the witness. In this way, by considering the geographical location information of the witness, the most suitable candidate can be presented.

[0056] The candidate presentation unit can analyze the witness's social media activity and propose a method for presenting candidates. For example, the candidate presentation unit can use AI to analyze the witness's social media activity and propose a method for presenting candidates. Based on the social media activity, the AI ​​can evaluate the reliability of the candidates and propose a method for presenting them. For example, the AI ​​analyzes the witness's social media posts and evaluates the reliability of the candidates. Based on the witness's social media activity, the AI ​​confirms the consistency of the candidates. The AI ​​evaluates the relevance of the candidates by referring to the witness's social media activity. In this way, by analyzing the witness's social media activity, a method for presenting candidates can be proposed.

[0057] The database matching unit can optimize the matching algorithm by referring to past matching history during database matching. For example, the database matching unit can use AI to refer to past matching history and optimize the matching algorithm. Based on past matching history, the AI ​​can propose the optimal matching algorithm and verify the consistency of the matching. For example, the AI ​​proposes the optimal matching algorithm based on past matching history. The AI ​​analyzes past matching history and verifies the consistency of the matching. By referring to past matching history, the AI ​​improves the accuracy of the matching. In this way, the matching algorithm can be optimized by referring to past matching history.

[0058] The database matching unit can apply different matching methods depending on the category of the data being matched during database matching. For example, the database matching unit can use AI to apply different matching methods depending on the category of the data being matched. The AI ​​can select the optimal matching method based on the category of the data being matched and perform the matching. For example, if the data being matched is a person, the AI ​​applies a matching method specialized for people. If the data being matched is an object, the AI ​​applies a matching method specialized for objects. If the data being matched is a landscape, the AI ​​applies a matching method specialized for landscapes. By applying different matching methods depending on the category of the data being matched, more accurate database matching becomes possible.

[0059] The database matching unit can weight matches based on the submission date of the items to be matched during database matching. For example, the database matching unit can use AI to weight matches based on the submission date of the items to be matched. The AI ​​can weight matches based on the submission date of the items to be matched. For example, if an item to be matched was recently submitted, the AI ​​will prioritize that match. If an item to be matched is old, the AI ​​will lower its priority. The AI ​​dynamically adjusts the matching weight according to the submission date of the items to be matched. This allows for prioritizing more important items to be matched by weighting matches based on the submission date of the items to be matched.

[0060] The database matching unit can improve the accuracy of matching by referring to related literature during database matching. For example, the database matching unit can use AI to refer to related literature and improve matching accuracy. Based on the related literature, the AI ​​can evaluate the reliability of the matching and confirm the consistency of the matching. For example, the AI ​​evaluates the reliability of the matching based on the related literature. The AI ​​confirms the consistency of the matching by referring to the related literature. The AI ​​improves the accuracy of the matching by analyzing the related literature. Thus, by referring to related literature, the accuracy of matching can be improved.

[0061] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0062] The image generation unit can detect whether a witness is focusing on a specific part of the generated montage image and highlight that part. For example, if a witness is focusing on the eyes in an image, the image generation unit can highlight that part to create an image that reflects the witness's intent. Similarly, if a witness is focusing on the mouth in an image, highlighting that part can create a more accurate montage image. Furthermore, if a witness is focusing on the outline of an image, highlighting that part can create an image based on the witness's memory. In this way, highlighting the parts that the witness is focusing on can create a more accurate montage image.

[0063] The database matching unit can adjust the matching priority for the images to be matched based on the reliability of the eyewitness testimony. For example, if the reliability of the eyewitness testimony is high, the database matching unit can prioritize matching based on that testimony. If the reliability of the eyewitness testimony is low, the database matching unit can lower the priority of matching based on that testimony. Furthermore, if the reliability of the eyewitness testimony is moderate, the database matching unit can perform matching based on that testimony with moderate priority. By adjusting the matching priority based on the reliability of the eyewitness testimony, more accurate database matching becomes possible.

[0064] The image generation unit can evaluate the consistency of eyewitness testimony in the generated montage image and highlight the parts with high consistency. For example, if the eyewitness testimony is consistent, the image generation unit can highlight that part to generate an image based on the eyewitness's memory. Conversely, if the eyewitness testimony is inconsistent, the image generation unit can not highlight that part to generate an image based on the eyewitness's memory. Furthermore, by evaluating the consistency of eyewitness testimony, the reliability of the testimony can be improved. As a result, by evaluating the consistency of eyewitness testimony and highlighting the parts with high consistency, a more accurate montage image can be created.

[0065] The candidate presentation unit can present multiple montage images generated based on the witness's testimony, based on the witness's past selection history. For example, it can analyze the characteristics of images previously selected by the witness and present the most suitable candidate based on those characteristics. Furthermore, by referring to the witness's past selection history, the candidate presentation unit can present consistent candidates. In addition, by analyzing the witness's past selection history, the candidate presentation unit can present candidates that match the witness's preferences. This makes it easier for the witness to select the most suitable montage image by presenting the most suitable candidate based on their past selection history.

[0066] The testimony analysis unit can analyze eyewitness testimonies by examining the eyewitness's social media activity and assessing the reliability of the testimony. For example, it can analyze the eyewitness's social media posts to confirm the consistency of the testimony. It can also evaluate the reliability of the testimony based on the eyewitness's social media activity. Furthermore, by referring to the eyewitness's social media activity, it can assess the relevance of the testimony. In this way, the reliability of testimony can be improved by analyzing the eyewitness's social media activity.

[0067] The following briefly describes the processing flow for example form 1.

[0068] Step 1: The testimony analysis unit analyzes the testimony and extracts the necessary facial features. For example, it uses a generative AI to analyze the testimony and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze the testimony using natural language processing technology and extract the necessary facial features. Step 2: The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. For example, it uses a generation AI to generate a composite image that reflects hairstyle, eye shape, face shape, etc. The generation AI can generate a composite image based on the extracted features using image generation technology. Step 3: The fine-tuning section allows the witness to make adjustments based on the image generated by the image generation section. For example, by fine-tuning the details while viewing the generated image, a more accurate montage can be created. The fine-tuning section can also use AI for adjustments. Step 4: The candidate presentation unit presents multiple candidates based on the images fine-tuned by the fine-tuning unit. For example, if memory is vague, multiple candidates can be presented, and the witness can select the best one to create the optimal montage. The candidate presentation unit can also use AI to present candidates. Step 5: The database matching unit compares the image selected by the witness from the candidates presented by the candidate suggestion unit with the criminal database. For example, AI can be used to match the image with images in the database and identify individuals who may potentially match.

[0069] (Example of form 2) The composite sketch creation system according to an embodiment of the present invention is a system that efficiently creates composite sketch images using a generative AI, thereby improving the speed and accuracy of investigations. Conventional composite sketch creation is time-consuming, and the results are uncertain due to the ambiguity of witness testimonies. The present invention solves these problems with the following components. First, natural language processing is used to analyze the witness's testimony and extract the necessary facial features. The generative AI analyzes the testimony and extracts features such as hairstyle, eye shape, and face shape. Next, an image generation AI is used to create a composite sketch image. Based on the extracted features, the AI ​​generates a composite sketch that reflects hairstyle, eye shape, face shape, etc. Furthermore, it has a function that allows the witness to make fine adjustments based on the generated image to improve its completeness. The witness can create a more accurate composite sketch by making fine adjustments to the details while looking at the generated image. Also, if memory is vague, the AI ​​presents multiple candidates, and the witness can select the optimal composite sketch. The created composite sketch image is compared with a criminal database. The AI ​​compares it with images in the database and identifies individuals who may match. Furthermore, this system can also be used as an aid in character design, allowing designers and creators to gain ideas for character creation. This system improves the speed of investigations and makes the most of witness memories. Additionally, its ease of use in generating character design ideas has led to increased demand in the design market. Thus, the montage creation system can improve the speed and accuracy of investigations.

[0070] The montage creation system according to this embodiment comprises a testimony analysis unit, an image generation unit, a fine-tuning unit, a candidate presentation unit, and a database matching unit. The testimony analysis unit analyzes the testimony and extracts necessary facial features. The testimony analysis unit analyzes the testimony using, for example, a generation AI and extracts features such as hairstyle, eye shape, and face shape. The generation AI can analyze the testimony using natural language processing technology and extract necessary facial features. The image generation unit generates a montage image based on the features extracted by the testimony analysis unit. The image generation unit generates a montage that reflects hairstyle, eye shape, face shape, etc., using, for example, a generation AI. The generation AI can generate a montage image based on the extracted features using image generation technology. The fine-tuning unit allows the witness to make fine adjustments based on the image generated by the image generation unit. The fine-tuning unit allows the witness to make fine adjustments to the details while looking at the generated image, for example, to create a more accurate montage. The fine-tuning unit can also make fine adjustments using AI. The candidate presentation unit presents multiple candidates based on images that have been fine-tuned by the fine-tuning unit. For example, if memory is vague, the candidate presentation unit can present multiple candidates, allowing the witness to select the best one to create the optimal composite image. The candidate presentation unit can also use AI to present candidates. The database matching unit matches the image selected by the witness from the candidates presented by the candidate presentation unit against a criminal database. For example, the database matching unit can use AI to match images in the database and identify individuals who may match. As a result, the composite image creation system according to this embodiment can improve the speed and accuracy of investigations by analyzing testimony, generating composite images, allowing witnesses to make fine adjustments, presenting multiple candidates, and matching them against a database.

[0071] The testimony analysis unit analyzes testimonies and extracts necessary facial features. For example, the testimony analysis unit uses generative AI to analyze testimonies and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze testimonies using natural language processing technology and extract necessary facial features. Specifically, the testimony analysis unit inputs testimonies obtained from eyewitnesses as text data, and the generative AI analyzes this text data. The generative AI extracts keywords and phrases related to facial features from the testimonies and identifies facial features based on them. For example, it analyzes descriptions such as "long hair," "round eyes," and "thin nose," quantifies each feature, and stores it in a database. Furthermore, the generative AI understands the context of the testimonies and can integrate multiple testimonies in a way that does not contradict each other. As a result, the testimony analysis unit can efficiently and accurately analyze eyewitness testimonies and extract necessary facial features. The testimony analysis unit also has a function to evaluate the reliability of testimonies and calculates a reliability score based on the consistency and detail of the testimonies. As a result, the testimony analysis unit can prioritize the analysis of highly reliable testimonies and extract more accurate facial features.

[0072] The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. For example, the image generation unit uses a generation AI to generate a composite image that reflects features such as hairstyle, eye shape, and face shape. The generation AI can generate a composite image based on extracted features using image generation technology. Specifically, the image generation unit inputs facial feature data provided by the testimony analysis unit, and the generation AI combines the various facial parts based on that data to generate a composite image. The generation AI selects parts such as hairstyle, eye shape, nose shape, and mouth shape based on a vast database of pre-trained facial images to generate a natural-looking face image. Furthermore, the generation AI adjusts the placement and size of the facial parts to create an accurate composite image based on the witness's testimony. The image generation unit displays the generated composite image in real time for the witness to review. The image generation unit also saves the generated composite image for use in subsequent processing. This allows the image generation unit to quickly and accurately generate a composite image based on the feature data provided by the testimony analysis unit.

[0073] The fine-tuning unit allows the user to make adjustments based on the image generated by the image generation unit. For example, the fine-tuning unit allows the user to create a more accurate montage by making fine adjustments to the details while viewing the generated image. The fine-tuning unit can also perform fine adjustments using AI. Specifically, the fine-tuning unit provides an interface that allows the user to adjust details such as hairstyle, eye shape, nose shape, and mouth shape while viewing the generated montage image. Through the interface, the user can fine-tune the shape and position of each part to create a more accurate montage image. The fine-tuning unit can also automatically make adjustments based on the user's instructions using AI. For example, if the user instructs the AI ​​to "make the eyes a little bigger," the AI ​​will automatically adjust the shape of the eyes based on that instruction. Furthermore, the fine-tuning unit can reflect the user's feedback in real time and display the adjustment results immediately. This allows the fine-tuning unit to create a more accurate montage by allowing the user to make fine adjustments to the details based on the generated montage image.

[0074] The candidate presentation unit presents multiple candidates based on the image fine-tuned by the fine-tuning unit. For example, if the memory is vague, the candidate presentation unit can present multiple candidates, allowing the witness to select the best one to create the optimal montage. The candidate presentation unit can also present candidates using AI. Specifically, the candidate presentation unit generates multiple variations based on the montage image adjusted by the fine-tuning unit and presents them to the witness. The AI ​​generates variations in different angles and facial expressions based on the features of the fine-tuned image, allowing the witness to select. The witness can then select the one that best matches their memory from the presented candidates, creating the optimal montage. The candidate presentation unit can also further narrow down the candidates based on the witness's selection. For example, if the witness selects a specific candidate, the unit further refines the details based on that candidate and presents new candidates. In this way, the candidate presentation unit can assist in creating the optimal montage based on the witness's memory.

[0075] The database matching unit compares the image selected by the witness from the candidates presented by the candidate suggestion unit with the criminal database. The database matching unit can, for example, use AI to compare images in the database and identify potentially matching individuals. Specifically, the database matching unit receives a composite image selected by the witness and compares it with images in the criminal database. Using facial recognition technology, the AI ​​compares the composite image with images in the database to identify potentially matching individuals. The AI ​​analyzes facial features and patterns to perform highly accurate matching. The database matching unit displays the matching results in a list format, sorted in descending order of matching degree. Investigators can use this list to identify suspects and determine the direction of the investigation. Furthermore, the database matching unit updates the matching results in real time, allowing for rapid response when new data is added. This enables the database matching unit to quickly and accurately identify potentially matching individuals based on the composite image selected by the witness and compared with the criminal database.

[0076] The database matching unit can identify individuals who may potentially match by comparing images in the criminal database. For example, the database matching unit can use AI to match images in the database and identify individuals who may potentially match. The database matching unit can also use facial recognition algorithms to match images in the database and identify individuals who may potentially match. For example, the database matching unit can set a similarity threshold and identify individuals with a similarity above that threshold. This allows for the identification of individuals who may potentially match by comparing images in the criminal database.

[0077] The testimony analysis unit can analyze testimonies using generative AI and extract necessary facial features. For example, the testimony analysis unit uses generative AI to analyze testimonies and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze testimonies using natural language processing technology and extract necessary facial features. For example, the generative AI takes text data of testimonies as input and outputs features such as hairstyle, eye shape, and face shape. As a result, by analyzing testimonies with the generative AI, the necessary facial features can be accurately extracted.

[0078] The image generation unit can generate a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI. For example, the image generation unit generates a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI. The generation AI can generate a montage image based on extracted features using image generation technology. For example, the generation AI takes features such as hairstyle, eye shape, and face shape as input and outputs a montage image. In this way, by generating a montage that reflects hairstyle, eye shape, face shape, etc., using a generation AI, an accurate montage image can be created.

[0079] The fine-tuning unit can improve the quality of the montage by making adjustments based on the images generated by the witness. For example, the fine-tuning unit can create a more accurate montage by making fine adjustments to the details while the witness is looking at the generated images. The fine-tuning unit can also use AI to make adjustments. For example, the fine-tuning unit can have the witness select a specific part of the image, and the AI ​​will make adjustments to that part. This allows the witness to make adjustments based on the generated images, thereby improving the quality of the montage image.

[0080] The candidate presentation unit presents multiple candidates when memory is unclear, allowing the witness to choose. For example, by presenting multiple candidates when memory is unclear and allowing the witness to select, the optimal composite image can be created. The candidate presentation unit can also use AI to present candidates. For example, the unit generates multiple candidates based on the witness's testimony, and the witness selects from among them. This allows the witness to select the optimal composite image by presenting multiple candidates when memory is unclear.

[0081] The database matching unit can also be used as an aid in character design, allowing designers and creators to gain ideas for character creation. For example, the database matching unit can use AI to assist in character design. To obtain ideas for character creation, the database matching unit can refer to images within the database and provide relevant information. This allows designers and creators to gain ideas for character creation by utilizing the database matching unit as an aid.

[0082] The testimony analysis unit can estimate the emotions of witnesses and evaluate the reliability of their testimony based on those estimated emotions. For example, the testimony analysis unit can use generative AI to estimate the emotions of witnesses and evaluate the reliability of their testimony based on those estimated emotions. The generative AI can use an emotion recognition algorithm to estimate the emotions of witnesses and evaluate the reliability of their testimony based on those emotions. For example, if a witness is nervous, the generative AI will rate the reliability of their testimony low. If a witness is relaxed, the generative AI will rate the reliability of their testimony high. If a witness is excited, the generative AI will rate the reliability of their testimony moderately. This allows for more accurate testimony analysis by evaluating the reliability of testimony based on the emotions of witnesses.

[0083] The testimony analysis unit can improve the accuracy of its analysis by referring to the witness's past testimony history during the analysis process. For example, the testimony analysis unit uses a generative AI to refer to the witness's past testimony history and evaluate the reliability of the current testimony. The generative AI can verify the consistency of the testimony and improve its reliability based on the past testimony history. For example, the generative AI evaluates the reliability of the current testimony by referring to the testimony the witness has provided in the past. The generative AI verifies the consistency of the testimony based on the witness's past testimony history. The generative AI improves the reliability of the testimony by analyzing the witness's past testimony history. In this way, the accuracy of testimony analysis can be improved by referring to the witness's past testimony history.

[0084] The testimony analysis unit can apply different analysis algorithms depending on the content of the testimony during the analysis. For example, the testimony analysis unit can use a generative AI to apply different analysis algorithms depending on the content of the testimony. The generative AI can select the optimal analysis algorithm based on the content of the testimony and analyze the testimony. For example, if the eyewitness testimony is detailed, the generative AI will apply a detailed analysis algorithm. If the eyewitness testimony is ambiguous, the generative AI will apply an analysis algorithm suitable for ambiguous testimony. If the eyewitness testimony is contradictory, the generative AI will apply an analysis algorithm to resolve the contradictions. In this way, the accuracy of testimony analysis can be improved by applying different analysis algorithms depending on the content of the testimony.

[0085] The testimony analysis unit can estimate the emotions of witnesses and adjust the importance of their testimonies based on those estimated emotions. For example, the testimony analysis unit can use a generative AI to estimate the emotions of witnesses and adjust the importance of their testimonies based on those estimated emotions. The generative AI can use an emotion recognition algorithm to estimate the emotions of witnesses and adjust the importance of their testimonies based on those emotions. For example, if a witness is nervous, the generative AI will set the importance of their testimony low. If a witness is relaxed, the generative AI will set the importance of their testimony high. If a witness is excited, the generative AI will set the importance of their testimony to a moderate level. By adjusting the importance of testimonies based on the emotions of witnesses, more accurate testimony analysis becomes possible.

[0086] The testimony analysis unit can prioritize the analysis of highly relevant testimonies by considering the geographical location information of witnesses during the testimony analysis process. For example, the testimony analysis unit uses a generative AI to consider the geographical location information of witnesses and prioritize the analysis of highly relevant testimonies. The generative AI evaluates the relevance of testimonies based on the geographical location information of witnesses and can prioritize the analysis of highly relevant testimonies. For example, if a witness is close to the crime scene, the generative AI will prioritize the analysis of that testimony. If a witness is far from the crime scene, the generative AI will lower the priority of that testimony. The generative AI evaluates the relevance of testimonies based on the geographical location information of witnesses. In this way, by considering the geographical location information of witnesses, it is possible to prioritize the analysis of highly relevant testimonies.

[0087] The testimony analysis unit can analyze witnesses' social media activity and analyze relevant testimonies during the testimony analysis process. For example, the testimony analysis unit can use generative AI to analyze witnesses' social media activity and analyze relevant testimonies. Based on social media activity, the generative AI can evaluate the reliability of testimonies and analyze relevant testimonies. For example, the generative AI evaluates the reliability of testimonies by analyzing witnesses' social media posts. Based on witnesses' social media activity, the generative AI confirms the consistency of the testimonies. The generative AI evaluates the relevance of testimonies by referring to witnesses' social media activity. In this way, relevant testimonies can be analyzed by analyzing witnesses' social media activity.

[0088] The image generation unit can estimate the emotions of the witness and adjust the expression method of image generation based on the estimated emotions. For example, the image generation unit can use a generation AI to estimate the emotions of the witness and adjust the expression method of image generation based on the estimated emotions. The generation AI can use an emotion recognition algorithm to estimate the emotions of the witness and adjust the expression method of image generation based on those emotions. For example, if the witness is relaxed, the generation AI will generate an image with a soft expression. If the witness is tense, the generation AI will generate an image with a sharp expression. If the witness is excited, the generation AI will generate an image with a vivid expression. In this way, by adjusting the expression method of image generation based on the emotions of the witness, it is possible to generate more appropriate images.

[0089] The image generation unit can adjust the level of detail of an image based on the importance of the testimony during image generation. For example, the image generation unit can use a generation AI to adjust the level of detail of an image based on the importance of the testimony. The generation AI can adjust the level of detail of an image based on the importance of the testimony. For example, if the testimony is detailed, the generation AI generates a high-level image. If the testimony is ambiguous, the generation AI generates a low-level image. The generation AI dynamically adjusts the level of detail of the image according to the importance of the testimony. This allows for the generation of more accurate images by adjusting the level of detail of the image based on the importance of the testimony.

[0090] The image generation unit can apply different generation algorithms depending on the category of the testimony during image generation. For example, the image generation unit can use a generation AI to apply different generation algorithms depending on the category of the testimony. The generation AI can select the optimal generation algorithm based on the category of the testimony and generate an image. For example, if the testimony concerns a person, the generation AI applies a person-specific generation algorithm. If the testimony concerns an object, the generation AI applies an object-specific generation algorithm. If the testimony concerns a landscape, the generation AI applies a landscape-specific generation algorithm. In this way, by applying different generation algorithms depending on the category of the testimony, more appropriate images can be generated.

[0091] The image generation unit can estimate the emotions of the witness and adjust the length of the image based on the estimated emotions. For example, the image generation unit can use a generation AI to estimate the emotions of the witness and adjust the length of the image based on the estimated emotions. The generation AI can use an emotion recognition algorithm to estimate the emotions of the witness and adjust the length of the image based on those emotions. For example, if the witness is relaxed, the generation AI will generate a longer image. If the witness is tense, the generation AI will generate a shorter image. If the witness is excited, the generation AI will generate an image of medium length. In this way, by adjusting the length of the image based on the emotions of the witness, it is possible to generate a more appropriate image.

[0092] The image generation unit can determine image priority based on the timing of testimony submission during image generation. For example, the image generation unit uses a generation AI to determine image priority based on the timing of testimony submission. The generation AI can determine image priority based on the timing of testimony submission. For example, if testimony has been recently submitted, the generation AI will prioritize generating images based on that testimony. If the testimony is old, the generation AI will lower its priority. The generation AI dynamically adjusts the image generation priority according to the timing of testimony submission. This allows for the generation of images based on more important testimony by determining image priority based on the timing of testimony submission.

[0093] The image generation unit can adjust the order of images based on the relevance of the testimonies during image generation. For example, the image generation unit uses a generation AI to adjust the order of images based on the relevance of the testimonies. The generation AI can adjust the order of images based on the relevance of the testimonies. For example, if the testimonies are highly relevant, the generation AI will prioritize the generation of images based on those testimonies. If the testimonies are not highly relevant, the generation AI will lower the priority of those testimonies. The generation AI dynamically adjusts the order of image generation according to the relevance of the testimonies. As a result, by adjusting the order of images based on the relevance of the testimonies, it is possible to generate images that are more relevant.

[0094] The fine-tuning unit can estimate the witness's emotions and adjust the fine-tuning method based on the estimated emotions. For example, the fine-tuning unit can use AI to estimate the witness's emotions and adjust the fine-tuning method based on the estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the witness's emotions and adjust the fine-tuning method based on those emotions. For example, if the witness is relaxed, the AI ​​provides detailed fine-tuning options. If the witness is tense, the AI ​​provides simple fine-tuning options. If the witness is excited, the AI ​​provides moderate fine-tuning options. This allows for more appropriate fine-tuning by adjusting the fine-tuning method based on the witness's emotions.

[0095] The fine-tuning unit can select the optimal adjustment method by referring to the witness's past adjustment history during fine-tuning. For example, the fine-tuning unit uses AI to refer to the witness's past adjustment history and select the optimal adjustment method. Based on the past adjustment history, the AI ​​can propose the optimal adjustment method and verify the consistency of the adjustment. For example, the AI ​​proposes the optimal adjustment method based on the adjustments the witness has made in the past. The AI ​​analyzes the witness's past adjustment history and verifies the consistency of the adjustment. The AI ​​improves the accuracy of the adjustment by referring to the witness's past adjustment history. In this way, the optimal adjustment method can be selected by referring to the witness's past adjustment history.

[0096] The fine-tuning unit can customize the means of adjustment based on the witness's current living situation during the fine-tuning process. For example, the fine-tuning unit uses AI to consider the witness's current living situation and customize the means of adjustment. The AI ​​can dynamically customize the means of adjustment based on the witness's living situation. For example, if the witness is busy, the AI ​​provides quick adjustment options. If the witness is relaxed, the AI ​​provides detailed adjustment options. The AI ​​dynamically customizes the means of adjustment according to the witness's living situation. This allows for more appropriate adjustments by customizing the means of adjustment based on the witness's current living situation.

[0097] The fine-tuning unit can estimate the witness's emotions and determine the priority of fine-tuning based on those estimated emotions. For example, the fine-tuning unit can use AI to estimate the witness's emotions and determine the priority of fine-tuning based on those estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the witness's emotions and determine the priority of fine-tuning based on those emotions. For example, if the witness is relaxed, the AI ​​will prioritize detailed fine-tuning. If the witness is tense, the AI ​​will prioritize simple fine-tuning. If the witness is excited, the AI ​​will prioritize moderate fine-tuning. This allows for more appropriate fine-tuning by determining the priority of fine-tuning based on the witness's emotions.

[0098] The fine-tuning unit can select the optimal adjustment method by considering the geographical location information of witnesses during the fine-tuning process. For example, the fine-tuning unit uses AI to consider the geographical location information of witnesses and select the optimal adjustment method. The AI ​​can evaluate the relevance of adjustments based on the geographical location information of witnesses and select the optimal adjustment method. For example, if a witness is close to the crime scene, the AI ​​will prioritize their testimony in the adjustment. If a witness is far from the crime scene, the AI ​​will lower the priority of their testimony. The AI ​​evaluates the relevance of adjustments based on the geographical location information of witnesses. In this way, the optimal adjustment method can be selected by considering the geographical location information of witnesses.

[0099] The fine-tuning unit can analyze the witness's social media activity during fine-tuning and propose adjustment methods. For example, the fine-tuning unit can use AI to analyze the witness's social media activity and propose adjustment methods. Based on the social media activity, the AI ​​can evaluate the reliability of the adjustment and propose adjustment methods. For example, the AI ​​analyzes the witness's social media posts and evaluates the reliability of the adjustment. Based on the witness's social media activity, the AI ​​confirms the consistency of the adjustment. The AI ​​refers to the witness's social media activity and evaluates the relevance of the adjustment. In this way, by analyzing the witness's social media activity, adjustment methods can be proposed.

[0100] The candidate presentation unit can estimate the witness's emotions and adjust the method of presenting candidates based on the estimated emotions. For example, the candidate presentation unit can use AI to estimate the witness's emotions and adjust the method of presenting candidates based on the estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the witness's emotions and adjust the method of presenting candidates based on those emotions. For example, if the witness is relaxed, the AI ​​will present detailed candidates. If the witness is tense, the AI ​​will present simple candidates. If the witness is excited, the AI ​​will present moderate candidates. By adjusting the method of presenting candidates based on the witness's emotions, more appropriate candidates can be presented.

[0101] The candidate suggestion unit can suggest the most suitable candidate by referring to the witness's past selection history. For example, the candidate suggestion unit can use AI to refer to the witness's past selection history and suggest the most suitable candidate. Based on the past selection history, the AI ​​can propose the most suitable candidate and verify the consistency of the candidates. For example, the AI ​​suggests the most suitable candidate based on the candidates the witness has previously selected. The AI ​​analyzes the witness's past selection history and verifies the consistency of the candidates. By referring to the witness's past selection history, the AI ​​improves the accuracy of the candidates. In this way, the most suitable candidate can be suggested by referring to the witness's past selection history.

[0102] The candidate presentation unit can customize the method of presenting candidates based on the witness's current living situation. For example, the candidate presentation unit uses AI to consider the witness's current living situation and customize the method of presenting candidates. The AI ​​can dynamically customize the method of presenting candidates based on the witness's living situation. For example, if the witness is busy, the AI ​​will provide quick candidate presentations. If the witness is relaxed, the AI ​​will provide detailed candidate presentations. The AI ​​dynamically customizes the means of presenting candidates according to the witness's living situation. This makes it possible to present more appropriate candidates by customizing the method of presenting candidates based on the witness's current living situation.

[0103] The candidate presentation unit can estimate the witness's emotions and determine the priority of candidate presentations based on the estimated emotions. For example, the candidate presentation unit can use AI to estimate the witness's emotions and determine the priority of candidate presentations based on the estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the witness's emotions and determine the priority of candidate presentations based on those emotions. For example, if the witness is relaxed, the AI ​​will prioritize detailed candidate presentations. If the witness is tense, the AI ​​will prioritize simple candidate presentations. If the witness is excited, the AI ​​will prioritize moderate candidate presentations. This allows for more appropriate candidate presentations by determining the priority of candidate presentations based on the witness's emotions.

[0104] The candidate presentation unit can present the most suitable candidate by considering the geographical location information of the witness. For example, the candidate presentation unit can use AI to consider the geographical location information of the witness and present the most suitable candidate. The AI ​​can evaluate the relevance of the candidates based on the geographical location information of the witness and present the most suitable candidate. For example, if the witness is close to the crime scene, the AI ​​will prioritize presenting that witness's testimony as a candidate. If the witness is far from the crime scene, the AI ​​will lower the priority of that witness's testimony. The AI ​​evaluates the relevance of the candidate presentations based on the geographical location information of the witness. In this way, by considering the geographical location information of the witness, the most suitable candidate can be presented.

[0105] The candidate presentation unit can analyze the witness's social media activity and propose a method for presenting candidates. For example, the candidate presentation unit can use AI to analyze the witness's social media activity and propose a method for presenting candidates. Based on the social media activity, the AI ​​can evaluate the reliability of the candidates and propose a method for presenting them. For example, the AI ​​analyzes the witness's social media posts and evaluates the reliability of the candidates. Based on the witness's social media activity, the AI ​​confirms the consistency of the candidates. The AI ​​evaluates the relevance of the candidates by referring to the witness's social media activity. In this way, by analyzing the witness's social media activity, a method for presenting candidates can be proposed.

[0106] The database matching unit can estimate the witness's emotions and adjust the database matching method based on the estimated emotions. For example, the database matching unit can use AI to estimate the witness's emotions and adjust the database matching method based on the estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the witness's emotions and adjust the database matching method based on those emotions. For example, if the witness is relaxed, the AI ​​performs a detailed database matching. If the witness is tense, the AI ​​performs a simple database matching. If the witness is excited, the AI ​​performs a moderate database matching. By adjusting the database matching method based on the witness's emotions, more appropriate database matching becomes possible.

[0107] The database matching unit can optimize the matching algorithm by referring to past matching history during database matching. For example, the database matching unit can use AI to refer to past matching history and optimize the matching algorithm. Based on past matching history, the AI ​​can propose the optimal matching algorithm and verify the consistency of the matching. For example, the AI ​​proposes the optimal matching algorithm based on past matching history. The AI ​​analyzes past matching history and verifies the consistency of the matching. By referring to past matching history, the AI ​​improves the accuracy of the matching. In this way, the matching algorithm can be optimized by referring to past matching history.

[0108] The database matching unit can apply different matching methods depending on the category of the data being matched during database matching. For example, the database matching unit can use AI to apply different matching methods depending on the category of the data being matched. The AI ​​can select the optimal matching method based on the category of the data being matched and perform the matching. For example, if the data being matched is a person, the AI ​​applies a matching method specialized for people. If the data being matched is an object, the AI ​​applies a matching method specialized for objects. If the data being matched is a landscape, the AI ​​applies a matching method specialized for landscapes. By applying different matching methods depending on the category of the data being matched, more accurate database matching becomes possible.

[0109] The database matching unit can estimate the emotions of the witness and determine the priority of database matching based on the estimated emotions. For example, the database matching unit can use AI to estimate the emotions of the witness and determine the priority of database matching based on the estimated emotions. The AI ​​can use an emotion recognition algorithm to estimate the emotions of the witness and determine the priority of database matching based on those emotions. For example, if the witness is relaxed, the AI ​​will prioritize detailed database matching. If the witness is tense, the AI ​​will prioritize simple database matching. If the witness is excited, the AI ​​will prioritize moderate database matching. This allows for more appropriate database matching by determining the priority of database matching based on the emotions of the witness.

[0110] The database matching unit can weight matches based on the submission date of the items to be matched during database matching. For example, the database matching unit can use AI to weight matches based on the submission date of the items to be matched. The AI ​​can weight matches based on the submission date of the items to be matched. For example, if an item to be matched was recently submitted, the AI ​​will prioritize that match. If an item to be matched is old, the AI ​​will lower its priority. The AI ​​dynamically adjusts the matching weight according to the submission date of the items to be matched. This allows for prioritizing more important items to be matched by weighting matches based on the submission date of the items to be matched.

[0111] The database matching unit can improve the accuracy of matching by referring to related literature during database matching. For example, the database matching unit can use AI to refer to related literature and improve matching accuracy. Based on the related literature, the AI ​​can evaluate the reliability of the matching and confirm the consistency of the matching. For example, the AI ​​evaluates the reliability of the matching based on the related literature. The AI ​​confirms the consistency of the matching by referring to the related literature. The AI ​​improves the accuracy of the matching by analyzing the related literature. Thus, by referring to related literature, the accuracy of matching can be improved.

[0112] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0113] The testimony analysis unit can analyze the tone of the witness's voice and speech characteristics when analyzing eyewitness testimony, and evaluate the reliability of the testimony. For example, if the witness's voice trembles during the testimony, the reliability of the testimony can be evaluated as low. Conversely, if the witness speaks in a consistent tone during the testimony, the reliability of the testimony can be evaluated as high. Furthermore, if the speed at which the witness speaks changes during the testimony, the reliability of that portion of the testimony can be re-evaluated. In this way, by considering the tone of the witness's voice and speech characteristics, the reliability of the testimony can be evaluated more accurately.

[0114] The image generation unit can detect whether a witness is focusing on a specific part of the generated montage image and highlight that part. For example, if a witness is focusing on the eyes in an image, the image generation unit can highlight that part to create an image that reflects the witness's intent. Similarly, if a witness is focusing on the mouth in an image, highlighting that part can create a more accurate montage image. Furthermore, if a witness is focusing on the outline of an image, highlighting that part can create an image based on the witness's memory. In this way, highlighting the parts that the witness is focusing on can create a more accurate montage image.

[0115] The candidate presentation unit can present multiple montage images generated based on the witness's testimony in different orders depending on the witness's emotions. For example, if the witness is relaxed, the candidate presentation unit can prioritize detailed images. If the witness is tense, the candidate presentation unit can prioritize simple images. If the witness is agitated, the candidate presentation unit can prioritize images with moderate detail. By presenting images in different orders according to the witness's emotions, it makes it easier for the witness to select the most suitable montage image.

[0116] The database matching unit can adjust the matching priority for the images to be matched based on the reliability of the eyewitness testimony. For example, if the reliability of the eyewitness testimony is high, the database matching unit can prioritize matching based on that testimony. If the reliability of the eyewitness testimony is low, the database matching unit can lower the priority of matching based on that testimony. Furthermore, if the reliability of the eyewitness testimony is moderate, the database matching unit can perform matching based on that testimony with moderate priority. By adjusting the matching priority based on the reliability of the eyewitness testimony, more accurate database matching becomes possible.

[0117] The testimony analysis unit can estimate the witness's emotions when analyzing their testimony and adjust the importance of the testimony based on those emotions. For example, if the witness is tense while giving their testimony, the importance of their testimony can be set low. If the witness is relaxed, the importance of their testimony can be set high. If the witness is excited, the importance of their testimony can be set to a moderate level. By adjusting the importance of the testimony based on the witness's emotions, a more accurate testimony analysis becomes possible.

[0118] The image generation unit can evaluate the consistency of eyewitness testimony in the generated montage image and highlight the parts with high consistency. For example, if the eyewitness testimony is consistent, the image generation unit can highlight that part to generate an image based on the eyewitness's memory. Conversely, if the eyewitness testimony is inconsistent, the image generation unit can not highlight that part to generate an image based on the eyewitness's memory. Furthermore, by evaluating the consistency of eyewitness testimony, the reliability of the testimony can be improved. As a result, by evaluating the consistency of eyewitness testimony and highlighting the parts with high consistency, a more accurate montage image can be created.

[0119] The fine-tuning unit can estimate the witness's emotions when they fine-tune the generated montage image and suggest fine-tuning methods based on those emotions. For example, if the witness is relaxed, the fine-tuning unit can provide detailed fine-tuning options. If the witness is tense, the fine-tuning unit can provide simple fine-tuning options. If the witness is excited, the fine-tuning unit can provide moderate fine-tuning options. This allows for more appropriate fine-tuning by suggesting methods based on the witness's emotions.

[0120] The candidate presentation unit can present multiple montage images generated based on the witness's testimony, based on the witness's past selection history. For example, it can analyze the characteristics of images previously selected by the witness and present the most suitable candidate based on those characteristics. Furthermore, by referring to the witness's past selection history, the candidate presentation unit can present consistent candidates. In addition, by analyzing the witness's past selection history, the candidate presentation unit can present candidates that match the witness's preferences. This makes it easier for the witness to select the most suitable montage image by presenting the most suitable candidate based on their past selection history.

[0121] The database matching unit can estimate the emotions of the witness in relation to the image being matched, and adjust the matching method based on the estimated emotions. For example, if the witness is relaxed, the database matching unit can perform a detailed match. If the witness is tense, the database matching unit can perform a simple match. If the witness is excited, the database matching unit can perform a match with a moderate level of detail. By adjusting the matching method based on the witness's emotions, more appropriate database matching becomes possible.

[0122] The testimony analysis unit can analyze eyewitness testimonies by examining the eyewitness's social media activity and assessing the reliability of the testimony. For example, it can analyze the eyewitness's social media posts to confirm the consistency of the testimony. It can also evaluate the reliability of the testimony based on the eyewitness's social media activity. Furthermore, by referring to the eyewitness's social media activity, it can assess the relevance of the testimony. In this way, the reliability of testimony can be improved by analyzing the eyewitness's social media activity.

[0123] The following briefly describes the processing flow for example form 2.

[0124] Step 1: The testimony analysis unit analyzes the testimony and extracts the necessary facial features. For example, it uses a generative AI to analyze the testimony and extract features such as hairstyle, eye shape, and face shape. The generative AI can analyze the testimony using natural language processing technology and extract the necessary facial features. Step 2: The image generation unit generates a composite image based on the features extracted by the testimony analysis unit. For example, it uses a generation AI to generate a composite image that reflects hairstyle, eye shape, face shape, etc. The generation AI can generate a composite image based on the extracted features using image generation technology. Step 3: The fine-tuning section allows the witness to make adjustments based on the image generated by the image generation section. For example, by fine-tuning the details while viewing the generated image, a more accurate montage can be created. The fine-tuning section can also use AI for adjustments. Step 4: The candidate presentation unit presents multiple candidates based on the images fine-tuned by the fine-tuning unit. For example, if memory is vague, multiple candidates can be presented, and the witness can select the best one to create the optimal montage. The candidate presentation unit can also use AI to present candidates. Step 5: The database matching unit compares the image selected by the witness from the candidates presented by the candidate suggestion unit with the criminal database. For example, AI can be used to match the image with images in the database and identify individuals who may potentially match.

[0125] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0126] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0127] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0128] Each of the multiple elements described above, including the testimony analysis unit, image generation unit, fine-tuning unit, candidate presentation unit, and database matching unit, is implemented in at least one of the smart device 14 and the data processing device 12. For example, the testimony analysis unit is implemented by the control unit 46A of the smart device 14, which analyzes the witness's testimony and extracts the necessary facial features. The image generation unit is implemented by the identification processing unit 290 of the data processing device 12, which generates a montage image based on the extracted features. The fine-tuning unit is implemented by the control unit 46A of the smart device 14, which allows the witness to make fine adjustments based on the generated image. The candidate presentation unit is implemented by the identification processing unit 290 of the data processing device 12, which presents multiple candidates. The database matching unit is implemented by the identification processing unit 290 of the data processing device 12, which matches the data against a criminal database. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0129] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0130] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0131] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0132] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0133] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0134] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0135] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0136] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0137] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0138] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0139] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0140] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0141] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0142] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0143] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0144] Each of the multiple elements described above, including the testimony analysis unit, image generation unit, fine-tuning unit, candidate presentation unit, and database matching unit, is implemented in at least one of the smart glasses 214 and the data processing unit 12. For example, the testimony analysis unit is implemented by the control unit 46A of the smart glasses 214, which analyzes the witness's testimony and extracts the necessary facial features. The image generation unit is implemented by the identification processing unit 290 of the data processing unit 12, which generates a montage image based on the extracted features. The fine-tuning unit is implemented by the control unit 46A of the smart glasses 214, which allows the witness to make fine adjustments based on the generated image. The candidate presentation unit is implemented by the identification processing unit 290 of the data processing unit 12, which presents multiple candidates. The database matching unit is implemented by the identification processing unit 290 of the data processing unit 12, which matches the data against a criminal database. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0145] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0146] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0147] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0148] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0149] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0150] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0151] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0152] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0153] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0154] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0155] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0156] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0157] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0158] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0159] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0160] Each of the multiple elements described above, including the testimony analysis unit, image generation unit, fine-tuning unit, candidate presentation unit, and database matching unit, is implemented in at least one of the headset terminal 314 and the data processing unit 12. For example, the testimony analysis unit is implemented by the control unit 46A of the headset terminal 314, which analyzes the witness's testimony and extracts the necessary facial features. The image generation unit is implemented by the identification processing unit 290 of the data processing unit 12, which generates a montage image based on the extracted features. The fine-tuning unit is implemented by the control unit 46A of the headset terminal 314, which allows the witness to make fine adjustments based on the generated image. The candidate presentation unit is implemented by the identification processing unit 290 of the data processing unit 12, which presents multiple candidates. The database matching unit is implemented by the identification processing unit 290 of the data processing unit 12, which matches the data against a criminal database. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0161] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0162] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0163] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0164] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0165] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0166] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0167] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0168] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0169] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0170] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0171] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0172] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0173] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0174] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0175] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0176] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0177] Each of the multiple elements described above, including the testimony analysis unit, image generation unit, fine-tuning unit, candidate presentation unit, and database matching unit, is implemented in at least one of the robot 414 and the data processing unit 12. For example, the testimony analysis unit is implemented by the control unit 46A of the robot 414, which analyzes the witness's testimony and extracts the necessary facial features. The image generation unit is implemented by the identification processing unit 290 of the data processing unit 12, which generates a montage image based on the extracted features. The fine-tuning unit is implemented by the control unit 46A of the robot 414, which allows the witness to make fine adjustments based on the generated image. The candidate presentation unit is implemented by the identification processing unit 290 of the data processing unit 12, which presents multiple candidates. The database matching unit is implemented by the identification processing unit 290 of the data processing unit 12, which matches the data against a criminal database. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0178] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0179] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0180] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0181] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0182] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0183] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0184] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0185] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0186] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0187] 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.

[0188] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0189] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0190] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0191] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0192] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0193] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0194] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0195] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0196] (Note 1) The testimony analysis unit analyzes the testimonies and extracts the necessary facial features, An image generation unit generates a montage image based on the features extracted by the aforementioned testimony analysis unit, A fine-tuning unit in which the witness makes adjustments based on the image generated by the image generation unit, A candidate presentation unit presents multiple candidates based on the image fine-tuned by the aforementioned fine-tuning unit, The system includes a database matching unit that compares an image selected by a witness from the candidates presented by the candidate presentation unit with a criminal database. A system characterized by the following features. (Note 2) The aforementioned database matching unit The images are matched against those in the criminal database to identify potential matches. The system described in Appendix 1, characterized by the features described herein. (Note 3) The aforementioned testimony analysis unit, The AI ​​generates the testimonies and extracts the necessary facial features. The system described in Appendix 1, characterized by the features described herein. (Note 4) The image generation unit, The AI ​​generates a montage that reflects hairstyle, eye shape, face shape, etc. The system described in Appendix 1, characterized by the features described herein. (Note 5) The aforementioned fine adjustment section is Witnesses make fine adjustments based on the generated images to improve their quality. The system described in Appendix 1, characterized by the features described herein. (Note 6) The candidate presentation unit, If the memory is vague, multiple options are presented, and the witness chooses. The system described in Appendix 1, characterized by the features described herein. (Note 7) The aforementioned database matching unit It can also be used as an aid in character design, and designers and creators can use it to get ideas for character creation. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned testimony analysis unit, We estimate the emotions of the witnesses and evaluate the reliability of their testimonies based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned testimony analysis unit, When analyzing testimonies, we improve the accuracy of the analysis by referring to the witness's past testimony history. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned testimony analysis unit, When analyzing testimonies, different analysis algorithms are applied depending on the content of the testimony. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned testimony analysis unit, Estimate the emotions of the witnesses and adjust the importance of their testimonies based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned testimony analysis unit, When analyzing testimonies, the geographical location of the witnesses is taken into consideration, and highly relevant testimonies are prioritized for analysis. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned testimony analysis unit, During the analysis of testimonies, the social media activity of witnesses is analyzed to identify relevant testimonies. The system described in Appendix 1, characterized by the features described herein. (Note 14) The image generation unit, The system estimates the emotions of the witnesses and adjusts the representation of the generated images based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 15) The image generation unit, When generating images, adjust the level of detail in the images based on the importance of the testimony. The system described in Appendix 1, characterized by the features described herein. (Note 16) The image generation unit, When generating images, different generation algorithms are applied depending on the category of the testimony. The system described in Appendix 1, characterized by the features described herein. (Note 17) The image generation unit, The system estimates the emotions of the witnesses and adjusts the length of the images based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 18) The image generation unit, When generating images, the priority of images is determined based on when the testimonies were submitted. The system described in Appendix 1, characterized by the features described herein. (Note 19) The image generation unit, When generating images, adjust the order of images based on the relevance of the testimonies. The system described in Appendix 1, characterized by the features described herein. (Note 20) The aforementioned fine adjustment section is Estimate the witness's emotions and adjust the fine-tuning method based on the estimated witness emotions. The system described in Appendix 1, characterized by the features described herein. (Note 21) The aforementioned fine adjustment section is During fine-tuning, the optimal adjustment method is selected by referring to the witness's past adjustment history. The system described in Appendix 1, characterized by the features described herein. (Note 22) The aforementioned fine adjustment section is During fine-tuning, the means of adjustment are customized based on the witness's current living situation. The system described in Appendix 1, characterized by the features described herein. (Note 23) The aforementioned fine adjustment section is The system estimates the emotions of the witnesses and determines the priority of fine-tuning based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 24) The aforementioned fine adjustment section is During fine-tuning, the optimal adjustment method is selected by considering the geographical location information of the witnesses. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned fine adjustment section is During fine-tuning, we analyze the social media activity of witnesses and suggest ways to make adjustments. The system described in Appendix 1, characterized by the features described herein. (Note 26) The candidate presentation unit, The system estimates the witness's emotions and adjusts the candidate presentation method based on the estimated witness emotions. The system described in Appendix 1, characterized by the features described herein. (Note 27) The candidate presentation unit, When presenting candidates, the system refers to the witness's past selection history to suggest the most suitable candidate. The system described in Appendix 1, characterized by the features described herein. (Note 28) The candidate presentation unit, When presenting candidates, customize the presentation method based on the witness's current living situation. The system described in Appendix 1, characterized by the features described herein. (Note 29) The candidate presentation unit, The system estimates the emotions of the witnesses and determines the priority of candidate presentations based on the estimated emotions of the witnesses. The system described in Appendix 1, characterized by the features described herein. (Note 30) The candidate presentation unit, When presenting candidates, the system will consider the geographical location information of the witnesses to present the most suitable candidates. The system described in Appendix 1, characterized by the features described herein. (Note 31) The candidate presentation unit, When presenting candidates, we analyze the social media activity of witnesses and propose methods for presenting candidates. The system described in Appendix 1, characterized by the features described herein. (Note 32) The aforementioned database matching unit The system estimates the emotions of the witnesses and adjusts the database matching method based on the estimated emotions of the witnesses. The system described in Appendix 1, characterized by the features described herein. (Note 33) The aforementioned database matching unit During database matching, the matching algorithm is optimized by referring to past matching history. The system described in Appendix 1, characterized by the features described herein. (Note 34) The aforementioned database matching unit When matching databases, different matching methods are applied depending on the category being matched. The system described in Appendix 1, characterized by the features described herein. (Note 35) The aforementioned database matching unit The system estimates the emotions of the witnesses and determines the priority of database matching based on the estimated emotions of the witnesses. The system described in Appendix 1, characterized by the features described herein. (Note 36) The aforementioned database matching unit When matching databases, weighting of the matching is performed based on the submission date of the items being matched. The system described in Appendix 1, characterized by the features described herein. (Note 37) The aforementioned database matching unit When matching databases, the accuracy of the matching is improved by referring to related literature being matched. The system described in Appendix 1, characterized by the features described herein. [Explanation of symbols]

[0197] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. The testimony analysis unit analyzes the testimonies and extracts the necessary facial features, An image generation unit generates a montage image based on the features extracted by the aforementioned testimony analysis unit, A fine-tuning unit in which the witness makes adjustments based on the image generated by the image generation unit, A candidate presentation unit presents multiple candidates based on the image fine-tuned by the aforementioned fine-tuning unit, The system includes a database matching unit that compares an image selected by a witness from the candidates presented by the candidate presentation unit with a criminal database. A system characterized by the following features.

2. The aforementioned database matching unit The images are matched against those in the criminal database to identify potential matches. The system according to feature 1.

3. The aforementioned testimony analysis unit, The AI ​​generates data to analyze testimonies and extract necessary facial features. The system according to feature 1.

4. The image generation unit, The AI ​​generates a montage that reflects hairstyle, eye shape, face shape, etc. The system according to feature 1.

5. The aforementioned fine adjustment section is Witnesses make fine adjustments based on the generated images to improve their quality. The system according to feature 1.

6. The candidate presentation unit, If the memory is vague, multiple options are presented, and the witness chooses. The system according to feature 1.

7. The aforementioned database matching unit It can also be used as an aid in character design, and designers and creators can use it to get ideas for character creation. The system according to feature 1.

8. The aforementioned testimony analysis unit, We estimate the emotions of the witnesses and evaluate the reliability of their testimonies based on those estimated emotions. The system according to feature 1.

9. The aforementioned testimony analysis unit, When analyzing testimonies, we improve the accuracy of the analysis by referring to the witness's past testimony history. The system according to feature 1.

10. The aforementioned testimony analysis unit, When analyzing testimonies, different analysis algorithms are applied depending on the content of the testimony. The system according to feature 1.

Images