Abnormality handling device and abnormality handling method

Abstract

The present invention appropriately handles an abnormality occurring in an abnormality determination target object.　An abnormality handling device 10 is provided with: a determination unit 11 that acquires information on an image obtained by imaging an abnormality determination target object, and determines an abnormality occurring in the abnormality determination target object on the basis of the acquired information on the image; a decision unit 12 that decides a handling policy for the abnormality occurring in the abnormality determination target object on the basis of the determination by the determination unit 11; and a generation unit 13 that generates a prompt to cause a generative AI model 21 to generate a handling policy explanation text for explaining the handling policy decided by the decision unit 12 to an abnormality handler.

Description

Abnormality response device and abnormality response method

[0001] The present invention relates to an abnormality response device and an abnormality response method used to respond to an abnormality occurring in a subject of abnormality judgment.

[0002] 2. Description of the Related Art Conventionally, a technique has been proposed for determining whether or not an abnormality has occurred on a road from an image of the road (see, for example, Patent Document 1).

[0003] JP 2017-107429 A

[0004] When road administrators, such as local governments, notice an abnormality in a road, for example, when there is a crack in the road, they need to take appropriate measures. In this regard, simply determining abnormalities from images, as in the prior art, is not necessarily sufficient.

[0005] One embodiment of the present invention has been made in consideration of the above, and aims to provide an abnormality response device and an abnormality response method that can appropriately respond to abnormalities occurring in an object to be determined as abnormal.

[0006] In order to achieve the above object, an anomaly response device according to one embodiment of the present invention comprises a judgment unit that acquires information relating to an image obtained by capturing an image of a target for anomaly detection and judges an abnormality occurring in the target for anomaly detection from the information relating to the acquired image, a decision unit that decides a response policy for the abnormality occurring in the target for anomaly detection in accordance with the judgment by the judgment unit, and a generation unit that generates a prompt to cause a generative AI model to generate a response policy explanation sentence that explains the response policy decided by the decision unit to a person responding to the abnormality.

[0007] In an anomaly response device according to one embodiment of the present invention, an anomaly is determined from information related to an image, a response policy for the anomaly is determined based on the determination, and a prompt is generated to cause a generative AI model to generate a response policy explanatory text that explains the determined response policy to a person responding to the anomaly. By using the prompt, the generative AI model can generate the response policy explanatory text. The generated response policy explanatory text allows a person to respond appropriately to the anomaly. Therefore, the anomaly response device according to one embodiment of the present invention can appropriately respond to an anomaly occurring in a target for anomaly determination.

[0008] Incidentally, one embodiment of the present invention can be described not only as an invention of an abnormality response device as described above, but also as an invention of an abnormality response method as described below. These are essentially the same inventions, just in different categories, and have similar functions and effects.

[0009] That is, an anomaly response method according to one embodiment of the present invention comprises a determination step in which an anomaly response device acquires information relating to an image obtained by capturing an image of a target for anomaly detection and determines an abnormality occurring in the target for anomaly detection from the information relating to the acquired image; a decision step in which the anomaly response device determines a response policy for the abnormality occurring in the target for anomaly detection based on the determination made in the determination step; and a generation step in which the anomaly response device generates a prompt to cause a generation AI model to generate a response policy explanation sentence that explains the response policy decided in the decision step to a person responding to the abnormality.

[0010] According to the present invention, it is possible to appropriately respond to an abnormality occurring in an object to be determined as abnormal.

[0011] 1 is a diagram illustrating a configuration of an abnormality handling device according to an embodiment of the present invention; FIG. 2 is a diagram illustrating an example of a prompt generated by an abnormality handling device; FIG. 3 is a flowchart illustrating an abnormality handling method which is a process executed by an abnormality handling device according to an embodiment of the present invention; and FIG. 4 is a diagram illustrating a hardware configuration of an abnormality handling device according to an embodiment of the present invention.

[0012] Hereinafter, an embodiment of an abnormality handling device and an abnormality handling method according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are given the same reference numerals and duplicated explanations will be omitted.

[0013] FIG. 1 shows an anomaly response device 10 according to this embodiment. The anomaly response device 10 is a device used to respond to an anomaly when an anomaly occurs in an object of anomaly detection. The object of anomaly detection is, for example, infrastructure in a city, such as a road managed by a local government. The anomaly to be detected is, for example, something that interferes with the proper use of the object of anomaly detection. If the object of anomaly detection is a road, the anomaly may be, for example, a crack or pothole in the road surface, or damage to the object, such as a broken guardrail. The anomaly may also be the presence of a hazardous object, such as weeds (e.g., weeds growing onto the sidewalk) or a fallen object. The anomaly may also be a vehicle accident on the road.

[0014] When the target of abnormality determination is one of the above, the abnormality response device 10 may be used by a manager of the target, such as a local government, to respond to an abnormality occurring in the target of abnormality determination. For example, the abnormality response device 10 may be used for road safety management by the local government.

[0015] The determination of whether an abnormality has occurred in the object of abnormality determination is performed based on an image. Therefore, the object of abnormality determination may be anything other than those described above, as long as an image can be acquired. Furthermore, the abnormality to be determined may be any abnormality, as long as it can be determined from an image.

[0016] Specifically, the abnormality response device 10 is used to generate a response policy explanatory text that explains the response policy for an abnormality occurring in the object of abnormality determination to a person responsible for responding to the abnormality (for example, a local government that manages the object of abnormality determination, the police, a road management company, or the Ministry of Land, Infrastructure, Transport and Tourism (a national organization)). By referring to the response policy explanatory text, the person responsible for responding to the abnormality can appropriately respond to the abnormality occurring in the object of abnormality determination.

[0017] The generation of the response policy explanation sentence itself is performed using a generative AI (artificial intelligence) model. The generative AI model is a model generated by machine learning. The generative AI model is a model that can generate content in response to a prompt input according to any one or a combination of the instructions, context, question, and output format indicated by the prompt, and return that content as response information. The prompt can also include input information, in which case the generative AI model generates response information targeted at the input information. The generative AI model may be, for example, an interactive AI model that includes an LLM (large-scale language model) and a user interface (UI) for interaction with the user, and enables text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, PaLM2, etc.

[0018] A prompt is a set of instructions or information input to a generative AI model. The prompt may include initial information, parameters, questions, etc., for the generative AI model to execute a specific task. A prompt is information indicating an instruction or question entered by a user in an interactive system, such as a dialogue with a generative AI model or a command line interface (CLI). The prompt may express, through text, for example, the command to be executed by the interactive AI model, the task to be executed by the interactive AI model, the background / context (e.g., role, condition) that the interactive AI model should consider, the question to be answered by the interactive AI model, and the output format of the response information from the interactive AI model. The prompt may also include input information that is the target of the command / task executed by the interactive AI model. Examples of such input information include data files with file names containing a predetermined extension, such as text data, image data, application-related data, audio data, video data, and still image data. Application-related data is data such as document data, table data, and graph data that can be processed by a default application program.

[0019] In this embodiment, the generative AI model 21 is realized by an AI server (server device) 20. The generative AI model 21 may be stored in the AI ​​server 20, or may be stored in another device connected to the AI ​​server 20 via a network and made available by the AI ​​server 20. Also, although the above describes an example of an LLM, other AI models may also be used. Note that the generative AI model 21 and the AI ​​server 20 may be similar to conventional models.

[0020] The anomaly response device 10 is configured by a computer such as a PC (personal computer) or a server device. The anomaly response device 10 may be configured by multiple computers. The anomaly response device 10 can transmit and receive information to and from other devices, including the AI ​​server 20, via a network to obtain information necessary to realize its functions. The anomaly response device 10 may be a Retrieval-Augmented Generation (RAG) system. In this case, the knowledge database searched by the RAG system may be implemented within the anomaly response device 10 or may be implemented in another device on a network (e.g., the cloud).

[0021] Next, a description will be given of the functions of the abnormality response device 10 according to this embodiment. As shown in FIG. 1, the abnormality response device 10 is configured to include a determination unit 11, a determination unit 12, and a generation unit 13.

[0022] The determination unit 11 is a functional unit that acquires information about an image obtained by capturing an image of a target for abnormality determination, and determines an abnormality occurring in the target for abnormality determination from the information about the acquired image. The determination unit 11 may acquire the image itself as information about the image.

[0023] The determination unit 11 acquires an image for determining an abnormality. The image is obtained by capturing (recording) an image of a target for abnormality determination (e.g., a road). The image may be a moving image. The image of the target for abnormality determination may be captured by any method. For example, the image of the target for abnormality determination may be captured using a camera provided on a vehicle traveling on the road that is the target for abnormality determination.

[0024] The determination unit 11 acquires an image by receiving an image transmitted from another device. Alternatively, the determination unit 11 acquires an image by accepting an image input operation by a user. The determination unit 11 may acquire an image by a method other than those described above.

[0025] The determination unit 11 may acquire, along with the image, location information indicating the location (place) where the image was captured and time information indicating the time when the image was captured. The location information and time information may be associated with the acquired image in advance (included in the image). The location information is, for example, information indicating the latitude and longitude of the location where the image was captured. If the target for abnormality determination is a road, the location information may indicate a position on the road. In this case, the location information may include at least one of the road name, direction of travel, kilometer posts, and surrounding facilities. By acquiring the location information and time information, it is possible to determine the location and time when an abnormality occurs in the target for abnormality determination.

[0026] The determination unit 11 determines (detects) an abnormality occurring in the target of abnormality determination from the acquired image. For example, the determination unit 11 determines whether a predetermined type of abnormality (e.g., a crack in a road) has occurred in the target of abnormality determination. Alternatively, the determination unit 11 determines the type of abnormality occurring in the target of abnormality determination. The determination unit 11 may perform a determination other than the above as long as it is a determination of an abnormality occurring in the target of abnormality determination. The determination from the image may be performed using conventional technology. For example, the determination may be performed using an image analysis AI that uses a learning model that has previously learned the relationship between learning images and events (anomalies).

[0027] The determination unit 11 may acquire information about the image other than the image itself as long as it is possible to determine an abnormality occurring in the target of the abnormality determination. For example, the determination unit 11 may acquire text information obtained by analyzing the image as the information. The determination unit 11 determines an abnormality occurring in the target of the abnormality determination from the acquired information. The determination from the information may be performed using conventional technology.

[0028] The determination unit 11 outputs information indicating the determination result to the determination unit 12. Furthermore, the determination unit 11 may output position information and time information related to the image used for the determination as information indicating the position and time related to the determination result to the determination unit 12. Note that if the determination unit 11 determines that no abnormality has occurred in the object of the abnormality determination, the subsequent processing may not be performed.

[0029] The determination unit 12 is a functional unit that determines a response policy for the abnormality occurring in the target of the abnormality determination, in accordance with the determination by the determination unit 11. The determination unit 12 may determine a response policy that includes a person who will respond to the abnormality.

[0030] The decision unit 12 decides a response policy for the abnormality, for example, as follows. The decision unit 12 receives information indicating the determination result from the determination unit 11. The decision unit 12 stores an abnormality, a response policy for the abnormality, and a person who will respond to the abnormality in association with each other. For example, the decision unit 12 stores a person who will respond to the abnormality and a response policy for the abnormality for each type of abnormality. The response policy for the abnormality is, for example, what the person who will respond to the abnormality should do. For example, in the case of a crack in a road, the person who will respond to the abnormality is the local government (prefecture) that is the road administrator, and the response policy for the abnormality is to repair the road.

[0031] When the information input from the determination unit 11 indicates that an abnormality has occurred in the target of anomaly determination, the determination unit 12 determines the abnormality response policy and the person who will respond to the abnormality that are associated with the abnormality in the pre-stored information as the abnormality response policy and the person who will respond to the abnormality that has occurred in the target of anomaly determination. Note that the number of people who will respond to the abnormality may be multiple. In other words, multiple people may respond to one abnormality. In this case, the abnormality response policy may be determined according to the person who will respond.

[0032] The determination unit 12 may input information indicating the location and time related to the determination result from the determination unit 11, and determine a policy for dealing with the abnormality and a person who will deal with the abnormality based on at least one of these. In this case, for example, the above-mentioned association information may be made to correspond to at least one of these.

[0033] Furthermore, the determination unit 12 may determine a response policy for an abnormality occurring in a target of the abnormality determination by a method other than the above, as long as the determination is made in accordance with the determination by the determination unit 11. The determination unit 12 outputs the determined information to the generation unit 13. The determination unit 12 also outputs the information input from the determination unit 11 to the generation unit 13.

[0034] The generation unit 13 is a functional unit that generates a prompt to cause the generation AI model 21 to generate a response policy explanatory sentence that explains the response policy decided by the decision unit 12 to a person responding to the anomaly. The generation unit 13 may include information indicating a predefined standard for responding to the anomaly in the prompt. The generation unit 13 may include an example of the response policy explanatory sentence in the prompt. The generation unit 13 may include information indicating the importance of the anomaly according to the person responding to the anomaly in the prompt.

[0035] The response policy explanatory text is a text that explains the response policy decided by the decision unit 12 to the person who will respond to the abnormality. The response policy explanatory text is intended to enable the person who will respond to the abnormality to appropriately respond to the abnormality occurring in the object of abnormality determination. The response policy explanatory text may include a text that indicates the abnormality determined by the determination unit 11. The response policy explanatory text may include a text that indicates how the person who will respond to the abnormality should specifically respond to the abnormality in accordance with the response policy decided by the decision unit 12. The response policy explanatory text may be based on a standard that is predefined for responding to an abnormality. The standard may be specific to each person who will respond to the abnormality. The standard may be, for example, a law (e.g., the Road Traffic Act) or a manual or rule that is predefined by a local government.

[0036] The response policy explanatory sentence is generated by the generation AI model 21. The generation unit 13 generates a prompt for causing the generation AI model 21 to generate the response policy explanatory sentence. Figure 2 shows an example of a prompt generated by the generation unit 13.

[0037] The generation unit 13 generates a prompt, for example, as follows: The generation unit 13 receives the information determined by the determination unit 12. The generation unit 13 stores rules and information (e.g., prompt templates) for generating a prompt in advance in a database or the like included in the anomaly response device 10, and generates a prompt from the rules and information and the information received from the determination unit 12.

[0038] The prompt includes information instructing the generation AI model 21 to generate a response policy explanation sentence. In the example shown in FIG. 2 , the information is a portion that reads, "Please create a response policy explanation sentence in accordance with the conditions below for the role veteran staff task." This portion is a standard phrase, and the generation unit 13 generates this portion using pre-stored information. As in the example shown in FIG. 2 , the information may include a portion that indicates to the generation AI model 21 what role (position) the response policy explanation sentence should take (in the above example, the portion that reads, "role veteran staff").

[0039] The prompt includes information indicating the conditions for generating a response policy explanatory sentence in the generation AI model 21. The conditions include information determined by the determination unit 12. The prompt includes information instructing the generation AI model 21 to take the conditions into consideration when generating a response policy explanatory sentence. In the example shown in FIG. 2 , the information is the following: "For report destinations that meet or exceed the conditions, please create a response policy explanatory sentence for each report destination while referring to examples. After creating the sentence, check whether the content of the created sentence is in line with the document examples. If it is not in line, please rewrite the sentence. If it is not in line, please tell us the name of the document that serves as the basis." This portion is a fixed phrase, and the generation unit 13 generates this portion using pre-stored information.

[0040] The prompt includes information indicating the person who will respond to the abnormality, that is, the person to whom the response policy explanation text will be reported. In the example shown in FIG. 2 , this information is the section "Report destination: road administrator, police." The generation unit 13 generates this section by embedding information indicating the person who will respond to the abnormality, input from the determination unit 12, at a preset position in a template text, which is pre-stored information.

[0041] The prompt includes information indicating a combination of the person who will respond to the abnormality, the determined abnormality, and a response policy. Furthermore, the combination may be further associated with information indicating a standard predefined for each person who will respond to the abnormality. In the example shown in FIG. 2 , the information is the following portion: "Document example: Road administrator (prefecture), road crack, do XX, A prefecture local government manual; Road administrator (prefecture), road crack, do YY, B prefecture local government manual." The generation unit 13 generates the portion by embedding each piece of information input from the determination unit 12 at a predetermined position in a template, which is pre-stored information.

[0042] The information in each line of the above section, "Road administrator (prefecture), road crack, do XX, A prefecture local government manual" and "Road administrator (prefecture), road crack, do YY, B prefecture local government manual," corresponds to one of the above combinations. In the above example, each piece of information is separated by "," and from left to right, it is information indicating the person who will respond to the abnormality, the determined abnormality, the response policy, and the norms.

[0043] In the above example, the information indicating the norms is the "Prefecture A Local Government Manual" and the "Prefecture B Local Government Manual," but the specific content of the norms (manuals) can be included in the prompt, for example, so that the generation AI model 21 can refer to the specific content of the norms (manuals) when generating a response policy explanation text.

[0044] Furthermore, as in the above example, when the generation AI model 21 generates a response policy explanatory text, it may check whether the generated response policy explanatory text complies with standards (such as local government manuals). If it does not comply, it may instruct the model to re-create (re-examine) the response policy explanatory text so that it complies.

[0045] In addition, the number of times that a norm (such as a local government manual) that was useful in detecting an inappropriate response policy explanation sentence was referenced may be counted, and the useful norm and its description section may be output to the generative AI model 21. This allows the user to identify the excellent description sections of the norm (such as a local government manual).

[0046] The prompt may include an example of a response policy explanation sentence. The example of the response policy explanation sentence may correspond to a combination of a person who will respond to the anomaly and the anomaly. In the example shown in FIG. 2 , the information is the following: "Please inform us of the example of the road administrator (prefecture), the crack in the road, the road name, the direction of travel, the kilometer posts, and surrounding facilities." This portion is a fixed phrase, and the generation unit 13 generates this portion using pre-stored information. As described above, this fixed phrase may correspond to a combination of a person who will respond to the anomaly and the anomaly.

[0047] The information in the line above, "Please inform us of the road administrator (prefecture), the cracked road, the road name, the direction of travel, the kilometer post, surrounding facilities, etc.", is an example of a response policy explanation sentence for a single anomaly and a combination of anomalies. In the above example, each piece of information is separated by "," and from left to right, the information shows an example of a response person for the anomaly, an anomaly, and a response policy explanation sentence.

[0048] The prompt includes information indicating the format of the output from the generation AI model 21. In the example shown in FIG. 2, this information is the section "output format abnormality content, response policy." In the above example, the format of the output from the generation AI model 21 is a list of abnormality content (determined abnormality) and response policy (response policy explanation text to be generated). This section is a fixed phrase, and the generation unit 13 generates this section using pre-stored information.

[0049] The generation unit 13 transmits the generated prompt to the AI ​​server 20 and inputs it to the generation AI model 21. The generation AI model 21 accepts the input of the prompt and generates a response policy explanatory sentence in accordance with the prompt. The generation AI model 21 transmits the generated response policy explanatory sentence to the anomaly response device 10 as a response to the transmitted prompt. The generation unit 13 receives and acquires the response policy explanatory sentence transmitted from the generation AI model 21.

[0050] The generation unit 14 may include information indicating the importance (e.g., effect) according to the responder (report recipient) in the prompt, and may have the importance taken into consideration when the generation AI model 21 generates a response policy explanation text. The information indicating the importance is, for example, information indicating the economic effect on the local government and citizen satisfaction. Alternatively, the information indicating the importance is information indicating civil engineering sales. Furthermore, the response policy explanation may be output according to the importance (in descending order of effect). In this case, the generation unit 13 may store in advance a correspondence relationship between the combination of the responder and the anomaly and the importance, and perform the above using the stored correspondence relationship.

[0051] In addition, the prompt input to the generation AI model 21 does not necessarily have to be the one described above, but may be any prompt that causes the generation AI model 21 to generate a response policy explanation sentence related to the response policy determined by the determination unit 12.

[0052] The acquired response policy explanatory text is used by the person responding to the anomaly to respond to the anomaly. For example, the anomaly response device 10 transmits the response policy explanatory text to the person responding to the anomaly. The person responding to the anomaly receives and refers to the response policy explanatory text, thereby responding to the anomaly. Furthermore, the transmission of the response policy explanatory text to the person responding to the anomaly does not have to be performed by the anomaly response device 10. The transmission of the response policy explanatory text to the person responding to the anomaly may be performed as an alert to the person responding to the anomaly. Furthermore, the information transmitted to the person responding to the anomaly may include location information and time information related to the image used in the determination. Furthermore, the response policy explanatory text may be used for purposes other than transmission to the person responding to the anomaly. The anomaly response device 10 may output the response policy explanatory text by a method other than the above, depending on the purpose of use of the response policy explanatory text.

[0053] Furthermore, the anomaly response device 10 only needs to generate the prompt, and the acquisition of the response policy explanation sentence from the generation AI model 21 using the prompt may be performed by a device other than the anomaly response device 10. In this case, the anomaly response device 10 only needs to output the generated prompt so that it can be used to acquire the response policy explanation sentence from the generation AI model 21. The above are the functions of the anomaly response device 10 according to this embodiment.

[0054] 3, an anomaly response method, which is a process executed by the anomaly response device 10 according to this embodiment (an operating method performed by the anomaly response device 10), will be described. In this process, first, the determination unit 11 acquires information about an image obtained by capturing an image of a target for anomaly detection (S01, determination step). Next, the determination unit 11 determines an anomaly occurring in the target for anomaly detection from the information about the acquired image (S02, determination step).

[0055] Next, the determination unit 12 determines a response policy for the abnormality occurring in the target of the abnormality determination in accordance with the determination by the determination unit 11 (S03, determination step). Next, the generation unit 13 generates a prompt to cause the generation AI model 21 to generate a response policy explanation sentence that explains the response policy determined by the determination unit 12 to the person responding to the abnormality (S04, generation step). Next, the generation unit 13 inputs the generated prompt to the generation AI model 21, and the response policy explanation sentence is acquired from the generation AI model 21 (S05). The acquired response policy explanation sentence is used by the person responding to the abnormality to respond to the abnormality. The above is the anomaly response method, which is processing executed by the anomaly response device 10 according to this embodiment.

[0056] In this embodiment, an abnormality is determined from information related to the image, a response policy for the abnormality is determined based on the determination, and a prompt is generated to cause the generation AI model 21 to generate a response policy explanatory text that explains the determined response policy to a person responding to the abnormality. By using the prompt, the generation AI model 21 can generate the response policy explanatory text. The generated response policy explanatory text allows a person to respond appropriately to the abnormality. Therefore, according to this embodiment, an abnormality occurring in a target for anomaly determination can be appropriately responded to.

[0057] For example, if an abnormality occurs in the target of abnormality judgment, information can be promptly shared with relevant parties. Furthermore, even if a local government or the like has a shortage of personnel to manage infrastructure, the abnormality can be appropriately dealt with. Furthermore, for example, regular patrols and support for local governments can be performed using cameras installed in vehicles and the generation AI model 21.

[0058] As in the present embodiment, the determination unit 11 may acquire the image itself as information about the image. This configuration makes it possible to appropriately and reliably determine whether an abnormality has occurred in the object of abnormality determination. As a result, it is possible to appropriately and reliably respond to the abnormality occurring in the object of abnormality determination. However, the determination unit 11 does not necessarily need to acquire the image itself; it is sufficient to acquire information about the image that can be used to determine whether an abnormality has occurred.

[0059] As in the present embodiment, the determination unit 12 may determine a response policy that includes a person who will respond to the abnormality. With this configuration, an abnormality occurring in a target for abnormality detection can be responded to by an appropriate person. However, the determination unit 12 does not necessarily need to determine a response policy that includes a person who will respond to the abnormality.

[0060] As in the present embodiment, the generation unit 13 may include information indicating the importance according to the responder in the prompt. According to this configuration, the generation AI model 21 can generate a response policy explanation sentence according to the importance. However, the generation unit 13 does not necessarily need to include information indicating the importance according to the responder in the prompt.

[0061] As in the present embodiment, the generation unit 13 may include information indicating a predefined rule for responding to an anomaly in the prompt. This configuration allows the generative AI model 21 to generate a response policy explanation sentence that conforms to the rule. However, the generation unit 13 does not necessarily need to include information indicating a predefined rule for responding to an anomaly in the prompt.

[0062] As in the present embodiment, the generation unit 13 may include an example of a response policy explanatory sentence in the prompt. According to this configuration, the generation AI model 21 can generate a response policy explanatory sentence corresponding to the example of the response policy explanatory sentence. However, the generation unit 13 does not necessarily have to include the example of the response policy explanatory sentence in the prompt.

[0063] In the above embodiment, the anomaly response device 10 is a PC or server device separate from the terminal of the user of the anomaly response device 10, but is not limited to this. For example, the anomaly response device 10 may be part of the terminal. As an example, a RAG system configured by the functional units of the anomaly response device 10 may be implemented in the terminal. In this case, the knowledge database searched by the RAG system may be implemented inside the terminal, or may be implemented in another device on a network (e.g., the cloud).

[0064] Furthermore, the anomaly response device 10 may have the functions of the AI ​​server 20. In this case, the configuration can be realized by installing an application that executes the functions of the AI ​​server 20 in the anomaly response device 10. In this way, the generative AI model 21 may be implemented on a network (e.g., the cloud) other than the AI ​​server 20.

[0065] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are directly or indirectly connected (e.g., wired, wireless, etc.) and these multiple devices. The functional block may also be realized by combining software with the single device or multiple devices.

[0066] Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0067] For example, the anomaly response device 10 according to an embodiment of the present disclosure may function as a computer that performs information processing according to the present disclosure. Figure 4 is a diagram illustrating an example of the hardware configuration of the anomaly response device 10 according to an embodiment of the present disclosure. The anomaly response device 10 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like. The hardware configuration of the AI ​​server 20 may also be as described herein.

[0068] In the following description, the term "apparatus" can be read as a circuit, a device, a unit, etc. The hardware configuration of the abnormality response apparatus 10 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0069] Each function of the abnormality response device 10 is realized by loading specified software (programs) onto hardware such as a processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0070] The processor 1001, for example, runs an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, each function of the abnormality response device 10 described above may be realized by the processor 1001.

[0071] Furthermore, the processor 1001 reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, each function of the anomaly response device 10 may be implemented by a control program stored in the memory 1002 and running on the processor 1001. While the above-described various processes have been described as being executed by one processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0072] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for performing information processing according to an embodiment of the present disclosure.

[0073] Storage 1003 is a computer-readable recording medium and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The storage medium provided in abnormality response device 10 may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0074] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0075] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0076] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0077] Furthermore, the abnormality response device 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0078] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0079] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0080] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0081] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0082] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0083] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0084] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0085] As used in this disclosure, the terms "system" and "network" are used interchangeably.

[0086] Furthermore, the information, parameters, etc. described in this disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information.

[0087] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0088] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0089] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0090] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0091] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0092] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0093] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0094] The anomaly response device and anomaly response method disclosed herein have the following configuration. [1] An anomaly response device including: a determination unit that acquires information about an image obtained by capturing an image of a target for anomaly detection and determines an anomaly occurring in the target for anomaly detection from the information about the acquired image; a determination unit that determines a response policy for the anomaly occurring in the target for anomaly detection based on the determination by the determination unit; and a generation unit that generates a prompt to cause a generative AI model to generate a response policy explanation sentence that explains the response policy decided by the determination unit to a person who will respond to the anomaly. [2] The anomaly response device described in [1], in which the determination unit acquires the image itself as information about the image. [3] The anomaly response device described in [1] or [2], in which the determination unit decides a response policy including a person who will respond to the anomaly. [4] The anomaly response device described in any of [1] to [3], in which the generation unit includes information indicating a predetermined standard for responding to an anomaly in the prompt. [5] The anomaly response device described in any of [1] to [4], in which the generation unit includes an example of a response policy explanation sentence in the prompt. [6] The anomaly response device according to any one of [1] to [5], wherein the generation unit includes in the prompt information indicating importance according to a person responding to the anomaly. [7] A method for responding to an anomaly, comprising: a determination step in which the anomaly response device acquires information about an image obtained by capturing an image of a target for anomaly detection and determines an anomaly occurring in the target for anomaly detection from the information about the acquired image; a determination step in which the anomaly response device determines a response policy for the anomaly occurring in the target for anomaly detection based on the determination in the determination step; and a generation step in which the anomaly response device generates a prompt to cause a generative AI model to generate a response policy explanatory sentence that explains the response policy determined in the determination step to a person responding to the anomaly.

[0095] 10...abnormality response device, 11...judgment unit, 12...decision unit, 13...generation unit, 20...AI server, 21...generated AI model, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.

Claims

1. An anomaly response device comprising: a judgment unit that acquires information about an image obtained by capturing an image of a target for anomaly detection and judges whether an abnormality has occurred in the target for anomaly detection from the information about the acquired image; a decision unit that decides a response policy for the abnormality occurring in the target for anomaly detection in accordance with the judgment by the judgment unit; and a generation unit that generates a prompt to cause a generative AI model to generate a response policy explanation sentence that explains the response policy decided by the decision unit to a person responsible for responding to the abnormality.

2. The anomaly response device according to claim 1, wherein the determination unit acquires the image itself as information relating to the image.

3. The abnormality response device according to claim 1, wherein the decision unit decides a response policy including a person who will respond to the abnormality.

4. The anomaly handling device according to claim 1, wherein the generating unit includes information indicating a predetermined standard for handling an anomaly in the prompt.

5. The anomaly response device according to claim 1, wherein the generation unit includes an example of a response policy explanation sentence in the prompt.

6. The anomaly response device according to claim 1, wherein the generation unit includes in the prompt information indicating the importance of the anomaly depending on the person who is responding to the anomaly.

7. A method for responding to an anomaly, comprising: a determination step in which an anomaly response device acquires information relating to an image obtained by capturing an image of a target for anomaly detection, and determines an abnormality occurring in the target for anomaly detection from the information relating to the acquired image; a decision step in which the anomaly response device determines a response policy for the abnormality occurring in the target for anomaly detection based on the determination made in the determination step; and a generation step in which the anomaly response device generates a prompt to cause a generative AI model to generate a response policy explanatory sentence explaining the response policy determined in the determination step to a person responding to the abnormality.

Images