Image processing apparatus, control method, and program

Abstract

To provide a technique for configuring optimal settings on image processing for detecting a defect more easily than before.SOLUTION: An image processing apparatus manages a result of detecting a defect on an image obtained by imaging an inspection target, as case information. The apparatus comprises acquisition means that acquires the case information according to first information including an image to be subjected to the defect detection and presentation means that presents the case information acquired by the acquisition means.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a technique for detecting abnormalities from an image of a test object.

Background Art

[0002] In the inspection of infrastructure, as a technique for detecting abnormalities from an image of a test object, there is a method of performing image processing using a learned model created by machine learning of AI (artificial intelligence) or deep learning which is a kind of machine learning. In this case, in order to accurately detect abnormalities, it is desirable to set an optimal model according to the image to be processed and adjust the parameters.

[0003] On the other hand, image processing for detecting abnormalities from an image of a test object requires a high-definition image, and the setting and processing are repeatedly performed until a desired result is obtained. Thus, since the image taken at the resolution required for image processing has a very large size, it takes time for processing and it is laborious to find an appropriate setting while repeatedly performing the setting and processing.

[0004] Patent Document 1 describes a method of recording in combination an example of image processing and a setting in which an image is trimmed, and presenting the example of image processing and the setting so that they can be selected.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In Patent Document 1, the user needs to select a target example from among the image processing examples presented. However, when it comes to image processing that performs deformation detection using a trained model and parameters, it can be difficult for the user to determine and select the optimal example simply by looking at the image to be processed and the detection results.

[0007] This invention has been made in view of the above problems, and its purpose is to realize a technology that makes it easier than before to set the optimal settings for image processing for detecting deformation. [Means for solving the problem]

[0008] To solve the above problems and achieve the objective, the present invention provides an image processing apparatus for managing the results of deformation detection performed on an image of an object to be inspected as case information, comprising: an acquisition means for acquiring the case information according to first information including the image on which the deformation detection is performed; and a presentation means for presenting the case information acquired by the acquisition means. [Effects of the Invention]

[0009] According to the present invention, it becomes easier than before to set the optimal settings for image processing for detecting deformations. [Brief explanation of the drawing]

[0010] [Figure 1] A block diagram showing the hardware configuration of the image processing device of this embodiment. [Figure 2] Functional block diagram of the image processing apparatus of this embodiment. [Figure 3] A flowchart illustrating the control process by the image processing apparatus of this embodiment. [Figure 4] This figure illustrates a UI screen for inputting deformation detection information in this embodiment. [Figure 5] A diagram illustrating the data structure of the case information in this embodiment. [Figure 6] A flowchart illustrating the process of presenting example information in this embodiment. [Figure 7]A diagram illustrating a UI screen for presenting case study information. [Figure 8] A diagram illustrating a UI screen for registering case information. [Figure 9] A diagram illustrating the data structure of case information, including newly registered case information. [Modes for carrying out the invention]

[0011] The embodiments will be described in detail below with reference to the attached drawings. Note that the following embodiments do not limit the invention as defined in the claims. While the embodiments describe multiple features, not all of these features are essential to the invention, and the features may be combined in any way. Furthermore, in the attached drawings, identical or similar configurations are given the same reference numerals, and redundant descriptions are omitted.

[0012] The following describes an embodiment in which the image processing device of the present invention is applied to a computer device used for inspecting infrastructure structures such as concrete structures, which are an example of objects to be inspected.

[0013] In this embodiment, a computer device operates as an image processing device and, based on deformation detection information (first information) including an image to be subjected to deformation detection, presents case information which is the result of a previously performed deformation detection process, thereby enabling the setting of image processing information (second information) used for image processing to perform deformation detection.

[0014] In this embodiment, the "inspection target" refers to concrete structures such as expressways, bridges, tunnels, and dams that are subject to inspection. The image processing device performs deformation detection processing to detect the presence and condition of deformations such as cracks using images captured by the user of the inspection target. "Deformation" refers to, for example, cracks, delamination, and spalling of concrete in the case of concrete structures. Other examples include efflorescence, exposed rebar, rust, water leakage, water dripping, corrosion, damage (defects), cold joints, precipitates, and honeycombing.

[0015] <Hardware Configuration> First, referring to FIG. 1, the hardware configuration of the image processing apparatus 100 according to the present embodiment will be described.

[0016] FIG. 1 is a block diagram showing the hardware configuration of the image processing apparatus 100 according to the present embodiment.

[0017] The processing of the embodiment described below may be realized by a single computer device, or each function may be distributed and realized by a plurality of computer devices as necessary. The plurality of computer devices are connected to be communicable with each other.

[0018] The image processing apparatus 100 includes a control unit 101, a nonvolatile memory 102, a work memory 103, a storage device 104, an input device 105, an output device 106, a network interface 107, and a system bus 108.

[0019] The control unit 101 includes arithmetic processing processors such as a CPU and an MPU that comprehensively control the entire image processing apparatus 100. The nonvolatile memory 102 is a ROM that stores programs, models, and parameters executed by the processor of the control unit 101. Here, the program is a program for executing the control process described later. Also, the model and parameters, which will be described later, are learned models used for image processing for detecting deformation from an image of an inspection target (hereinafter, detection image) and parameters used for learning processing. The work memory 103 is a RAM that temporarily stores programs and data supplied from an external device or the like.

[0020] The storage device 104 is an internal device such as a hard disk or memory card built into the image processing device 100, or an external device such as a hard disk or memory card detachably connected to the image processing device 100. The storage device 104 includes memory cards and hard disks made of semiconductor memory or magnetic disks. The storage device 104 also includes a storage medium consisting of a disk drive that reads / writes data to / from optical discs such as CDs, DVDs, and Blu-ray Discs.

[0021] The control unit 101 performs deformation detection processing to detect deformations from a detection image by image processing using a trained model and parameters. The trained model is created, for example, by machine learning or deep learning, a type of machine learning, in artificial intelligence (AI). The trained model can be configured as, for example, a neural network model. The training process may also be performed by a GPU (Graphics Processing Unit). A GPU is a processor capable of performing processing specialized for computer graphics calculations and has the processing power to perform matrix operations and other calculations necessary for the training process in a short time. The training process is not limited to a GPU; any circuit configuration that can perform matrix operations and other calculations necessary for a neural network is acceptable.

[0022] Furthermore, the trained model and parameters used for image processing to detect abnormalities from images of the object being inspected may be obtained from a cloud server connected to the network via the network interface 107. Alternatively, the detection images and parameters may be sent to the cloud server, and the detection results obtained by performing image processing (inference processing) using the trained model on the cloud server may be obtained as case information via the network interface 107.

[0023] The input device 105 is an operating component such as a mouse, keyboard, or touch panel that accepts user input and outputs operation instructions to the control unit 101. The output device 106 is a display device such as a display or monitor made of an LCD or organic EL, which displays data held by the image processing device 100 or data supplied from external devices. The network interface 107 is connected to a network such as the Internet or a LAN (Local Area Network) for communication. The system bus 108 connects the components 101 to 107 of the image processing device 100 so that data can be exchanged.

[0024] The non-volatile memory 102 or storage device 104 stores the operating system (OS), which is the basic software executed by the control unit 101, and applications that work in cooperation with the OS to realize advanced functions. In this embodiment, the non-volatile memory 102 or storage device 104 also stores applications that enable the image processing device 100 to perform image analysis processing to detect abnormalities from images of the object to be inspected, as described later.

[0025] The processing of the image processing device 100 in this embodiment is achieved by loading software provided by an application. The application is assumed to have software for utilizing the basic functions of the OS installed on the image processing device 100. The OS of the image processing device 100 may also have software for implementing the processing in this embodiment.

[0026] <Functional Configuration> Next, with reference to Figure 2, the functional blocks of the image processing apparatus 100 of this embodiment will be described.

[0027] Figure 2 is a functional block diagram of the image processing apparatus 100 of this embodiment.

[0028] The image processing device 100 comprises a receiving unit 201, a management unit 202, an acquisition unit 203, a display unit 204, a detection unit 205, and a registration unit 206. Each function of the image processing device 100 is composed of hardware and software. Note that each functional unit may be composed of one or more computer devices or server devices, and the system may be configured as a network-connected system.

[0029] The receiving unit 201 receives deformation detection information, including image information of the detection image, and image processing information used for image processing from an external server device or the like via the storage device 104, input device 105, or network interface 107.

[0030] The management unit 202 manages the registration, deletion, and updating of case information stored in the memory device 104. The case information is historical information regarding the settings and results of previously executed anomaly detection processes.

[0031] The acquisition unit 203 acquires case information from the management unit 202 based on the abnormality detection information received by the receiving unit 201.

[0032] The presentation unit 204 presents the case information acquired by the acquisition unit 203 to the user by displaying it, etc.

[0033] The detection unit 205 performs a process to detect deformations from the detection image based on the deformation detection information and image processing information received by the receiving unit 201.

[0034] The registration unit 206 registers the results of the abnormality detection performed by the detection unit 205 as case information in the management unit 202.

[0035] <Control Processing> Next, the control process of this embodiment will be described with reference to Figures 3 to 9.

[0036] Figure 3 is a flowchart showing the control process of the image processing device 100 in this embodiment.

[0037] The process shown in Figure 3 is realized by the control unit 101 of the image processing device 100 shown in Figure 1 loading the program stored in the non-volatile memory 102 into the work memory 103 and executing it, thereby controlling each component shown in Figure 1 and operating as each functional unit shown in Figure 2. Furthermore, the process shown in Figure 3 is started when the image processing device 100 receives an instruction from the input device 105 to start the deformation detection process.

[0038] In S301, the receiving unit 201 receives information for detecting deformation. In this case, the user, for example, operates the input device 105 to input the file name of the detection image into the image selection field 402 of the UI screen 401 shown in Figure 4. Based on the information entered into the image selection field 402, the receiving unit 201 receives the detection image stored in the storage device 104 via the management unit 202. In addition to the image selection field 402, the UI screen 401 shown in Figure 4 is also provided with a type input field 403 and a component input field 404 for inputting the type and components of the structure to be inspected, so the user can input the type and components of the structure to be inspected as deformation detection information. In this case, the receiving unit 201 receives information on the type and components of the structure to be inspected in addition to the detection image entered into the image selection field 402. Furthermore, in addition to the image selection field 402, type input field 403, and component input field 404, the UI screen 401 shown in Figure 4 is also provided with a user input field 405 and an organization input field 406 for inputting the user who entered the deformation detection information and the organization to which the user belongs. Users can input their own identification information and their organization's identification information as detection information. In this case, the receiving unit 201 receives the user's identification information and their organization's identification information in addition to the information entered in the image selection field 402, type input field 403, and component input field 404. Note that the method of inputting deformation detection information is just one example and is not limited to the input method using the UI screen 401 shown in Figure 4.

[0039] In S302, the acquisition unit 203 acquires case information stored in the storage device 104 via the management unit 202 based on the deformation detection information received by the receiving unit 201 in S301. Figure 5(a) illustrates the data structure of case information 501A stored in the storage device 104. In Figure 5(a), each row of case information 501A is history information regarding the settings and results of deformation detection processing performed in the past. The history information includes, for example, case ID 502, image 503, model 504, parameters 505, and detection result 506. Case ID 502 registers identification information assigned to each case. Image 503 registers the file name of the image used for detection. Model 504 and parameter 505 register the name of the trained model used for image processing of the image used for detection and the parameters used for the training process as image processing information. Detection result 506 registers the deformation information detected from the image used for detection.

[0040] Furthermore, the history information includes attribute information such as the type 507 and member 508 of the structure being inspected. In addition, the history information includes, as supplementary information, a disclosure range 509 indicating the range of users to whom the case information is made public, an appropriateness level 510 indicating the level of appropriateness of the deformation detection results, and a description field 511 for describing the case information. Note that the case information may pre-register information used in the training process of the learning model in the deformation detection process, or the description field 511 may be pre-registered with the correct values ​​and features of the image 503 and detection results 506 based on the information used in the training process of the learning model.

[0041] The acquisition unit 203 acquires information from all rows of the case information 501A shown in Figure 5(a), or information from rows corresponding to the deformation detection information received by the receiving unit 201 in S301, or information from rows selected based on the deformation detection information. For example, if the acquisition unit 203 receives the type of structure to be inspected, "bridge," and the member, "floor slab," as deformation detection information, it acquires case information 512 from the case information 501B shown in Figure 5(b), in which the type of structure to be inspected and the member match. Also, if the acquisition unit 203 receives the type of structure to be inspected, "bridge," and the member, "floor slab," as deformation detection information, in addition to the user ID "X" and organization "A," it acquires case information 513 from the case information 501B shown in Figure 5(b), in which the type of structure to be inspected and the member match, and the scope of disclosure includes the user ID "X." Furthermore, the acquisition unit 203 may acquire case information similar to the images and EXIF ​​information attached to the images that are registered in the case information 501B shown in Figure 5(b), as well as the detection images and EXIF ​​information attached to the detection images acquired as information for detecting abnormalities.

[0042] In S303, the presentation unit 204 presents to the user the detection image received by the receiving unit 201 and the case information acquired by the acquisition unit 203. Figure 5(c) illustrates a portion of the case information 501C acquired by the acquisition unit 203 in S302. The presentation process in S303 will be described later in Figure 6.

[0043] In S304, the detection unit 205 performs a process to detect deformations in the detection image received by the receiving unit 201, using the model and parameters of the case information selected by the user from the case information 501C presented by the presentation unit 204.

[0044] In S305, the registration unit 206 registers the results of the deformation detection performed by the detection unit 205 as case information. In this case, the user can register the case information by, for example, operating the input device 105 and entering the case information into the UI screen 801 shown in Figure 8. In the UI screen 801, the setting input field 802 allows input of the model and parameter settings used in the deformation detection process of S304, and the input information is displayed. The detection result field 803 allows input of deformation information from the deformation detection process results of S304, and the entered deformation information is displayed. The publication range field 804 displays options for the scope to which the registered case information will be published, which can be selected from a pull-down menu. The appropriateness field 804 displays options for the appropriateness of the deformation detection result of the registered case information, which can be selected from a pull-down menu. When the user operates the finish button 806, the case information entered into the UI screen 801 in Figure 8 is added to the case information 501A in Figure 5(a).

[0045] Figure 9 illustrates case information 901, which includes case information 902 added in S305. In case information 902, newly created identification information is registered as case ID 502. Also, in S301, the detection image received by the receiving unit 201 (the image entered in the image selection field 402 of the UI screen 401 in Figure 4), the type of structure to be inspected, and the members are registered as image 503, type 507, and member 508, respectively. Furthermore, in S304, the model and parameters used in the deformation detection process executed by the detection unit 205, and the deformation detection results are registered as model 504, parameter 505, and detection result 506, respectively. In addition, in S305, the disclosure range and appropriateness entered in the UI screen 801 in Figure 8 are registered as disclosure range 509 and appropriateness 510, respectively. Furthermore, if "None" is selected as the option for the publication scope field 804 in the UI screen 801 of Figure 8, the case information entered in the UI screen 801 of Figure 8 may be not registered.

[0046] <Processing for presenting case information> Next, with reference to Figure 6, the process of presenting case information in S303 of Figure 3 will be explained.

[0047] Figure 6 is a flowchart illustrating the process of presenting case information in S303 of Figure 3. Figure 7 shows an example of a UI screen for presenting case information.

[0048] In S601, the display unit 204 displays the case information acquired by the acquisition unit 203 in S302 of Figure 3 on the output device 106. The display unit 204 displays, for example, the case information list screen 700 shown in Figure 7(a). The case information list screen 700 displays a list of model names 701 and case examples 702 of the models that can be set in the deformation detection process executed by the detection unit 205 in S304 of Figure 3. For case examples 702, a case description 703 for each model and one or more case images 704 are displayed. The case description 703 and case image 704 display the case description and image for each model together. For example, the case description 703 displays the content written in the description field 511 of case information 515 in case information 501C shown in Figure 5(c) where the model name matches "Model 001". Similarly, in the case image 704, a thumbnail of the image registered in image 503 of case information 515, where the model name matches "Model 001" in the case information 501C shown in Figure 5(c), is displayed. Furthermore, in the case information list screen 700 shown in Figure 7(a), the user may select model name 701 to set the model used for the deformation detection process executed by the detection unit 205 in S304 of Figure 3.

[0049] In S602, the receiving unit 201 operates the input device 205, for example, on the model setting screen 705 shown in Figure 7(b) or the parameter setting screen 710 shown in Figure 7(cb), to receive instructions for setting a model or parameters used in the deformation detection process performed by the detection unit 205 in S304 of Figure 3.

[0050] In S603, the presentation unit 204 determines whether the setting instruction received by the receiving unit 201 indicates that the setting is complete. If the setting instruction received by the receiving unit 201 indicates that the setting is complete, the presentation unit 204 terminates the process and proceeds to S304 in Figure 3. If the setting instruction received by the receiving unit 201 does not indicate that the setting is complete, the presentation unit 204 proceeds to S604.

[0051] In S604, the presentation unit 204 determines whether the setting instruction received by the receiving unit 201 is a model setting. If the setting instruction received by the receiving unit 201 is a model setting, the presentation unit 204 proceeds to S605. If the setting instruction received by the receiving unit 201 is not a model setting, the presentation unit 204 proceeds to S606.

[0052] In S605, the display unit 204 displays case information of the model selected by the setting instruction (model setting) received by the receiving unit 201 in S602 on the output device 106. The display unit 204 displays, for example, the model setting screen 705 shown in Figure 7(b). The model setting screen 705 displays statistical quantities 706 for each structure to be inspected in which deformation detection processing has been performed using the selected model, as well as suitable examples 707 and unsuitable examples 708 indicating the suitability of the detection image. The statistical quantities 706 display a graph showing the frequency of each type of structure to be inspected that is registered in the type 507 of case information 515 in the case information 501C shown in Figure 5(c) where the model name matches "Model 001". The suitable examples 707 display the image 503 and detection result 506 of case information 516 in the case information 501C shown in Figure 5(c) where the model name matches "Model 001" and the detection result is registered in the image 503 and detection result 506. Similarly, in the case of unsuitable example 708, the image 503 and detection result 506 registered in case information 517, which matches the model name "Model 001" in the case information 501C shown in Figure 5(c), and has the lowest suitability level of 1, are displayed. Furthermore, by operating the parameter setting button 709 on the model setting screen 705, a UI screen for the user to set parameters is displayed.

[0053] In S606, the presentation unit 204 determines whether the setting instruction received by the receiving unit 201 is a parameter setting. If the setting instruction received by the receiving unit 201 is a parameter setting, the presentation unit 204 proceeds to S607. If the setting instruction received by the receiving unit 201 is not a parameter setting, the presentation unit 204 returns to S602.

[0054] In S607, the display unit 204 displays example information of the parameters set by the setting instruction (parameter setting) received by the receiving unit 201 in S602 for the model selected in S605 on the output device 106. In this case, the display unit 204 displays, for example, the parameter setting screen 710 shown in Figure 7(c). The parameter setting screen 710 displays multiple (3 types) sliders 711 that can be used to set parameters for each parameter, and an example of the detection result 712 for the set model and parameters. In the example in Figure 7(c), the parameter "3,3,3" is set by the slider 711. In this case, the example of the detection result 712 displays the detection result registered in the detection result 506 of the example information 516 shown in Figure 5(c), where the model name is "Model 001" and the parameters are "3,3,3". Next, when the parameter "5,3,3" is set by the slider 713, the detection result example 714 will display the detection result registered in detection result 506 of case information 518, which matches the model name "Model 001" and the parameter "5,3,3" in the case information 501C shown in Figure 5(c). If there are multiple case information entries with matching parameters in the case information 501C shown in Figure 5(c), multiple entries may be displayed. Furthermore, in the case information 501C shown in Figure 5(c), case information that matches the type or component of the structure to be inspected received by the receiving unit 201 in S301, or case information with the highest level of appropriateness, may be selected and displayed. In addition, if there is no case information with matching parameters in the case information 501C shown in Figure 5(c), the detection result of case information in which the parameter most similar to the parameter set on the parameter setting screen 710 in Figure 7(c) is registered among the parameters registered in parameter 505 of case information 501C may be displayed.

[0055] Furthermore, when the user operates the "Setting Complete" button 715 on the parameter setting screen 710 in Figure 7(c) using the input device 205, the presentation unit 204 determines in S603 that the parameter and model settings have been instructed to be complete and terminates the process.

[0056] Furthermore, as a method for presenting example detection results according to parameter settings, the example detection results may be presented in association with the slider for setting the parameters. For example, as shown in Figure 7(d), an example detection result 718 is displayed at the position corresponding to the parameter value 717 of the slider 716 for setting the parameters. In this case, highlighting 719 may be performed to indicate that case information corresponding to parameter value 717 exists. In addition, as shown in Figure 7(e), the parameter value 721 of the slider 720 for setting the parameters may be highlighted by coloring or other means 722. Moreover, the difference between the detection result corresponding to parameter value 721 and the detection result corresponding to a parameter value smaller than parameter value 721 may be extracted and the extracted area may be enlarged and highlighted 723.

[0057] Furthermore, the method for presenting case information in S303 in Figure 3 (S610 in Figure 6) is not limited to the case information list screen 700 in Figure 7(a); the model setting screen 705 in Figure 7(b) may be displayed as a list, or the parameter setting screen 710 in Figure 7(c) may also be displayed in S605 in Figure 6. In addition, one of the processes in S601, S605, or S607 in Figure 6 may be omitted, or a part of the UI screen may be hidden.

[0058] Furthermore, although this embodiment describes an example in which the acquisition unit 203 acquires case information and the presentation unit 204 presents it, the acquisition unit 203 may acquire case information and the presentation unit 204 may present the case information each time the receiving unit 201 receives deformation detection information, and this process may be repeated. As described above, according to this embodiment, by presenting case information corresponding to the deformation detection information input by the user, it becomes easier to set the optimal model and parameters than in the conventional method, and detection processing can be performed efficiently while maintaining detection accuracy. As a result, the reliability of deformation detection by image processing is improved, and inspection work can be made more efficient.

[0059] [Other embodiments] The present invention can also be realized by supplying a program that implements one or more functions of each embodiment to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. Furthermore, the present invention can also be realized by a circuit (e.g., an ASIC) that implements one or more functions.

[0060] The disclosures herein include the following image processing apparatus, control methods, and programs. [Configuration 1] In an image processing device that manages the results of anomaly detection performed on images of an object to be inspected as case information, An acquisition means for acquiring the case information in accordance with the first information, which includes an image used for detecting the deformation, An image processing apparatus characterized by having a presentation means for presenting case information acquired by the acquisition means. [Configuration 2] The image processing apparatus according to configuration 1, further comprising a receiving means for receiving the first information. [Configuration 3] The image processing apparatus according to configuration 2, characterized in that the first information includes an image used for detecting deformation and information about the structure to be inspected. [Structure 4] The image processing apparatus according to configuration 3, characterized in that the information of the structure to be inspected includes the type and components of the structure to be inspected. [Composition 5] The image processing apparatus according to any one of configurations 1 to 4, characterized in that the aforementioned case information includes the image on which the deformation was detected and second information used for image processing to perform the deformation detection. [Composition 6] The image processing apparatus according to configuration 5, characterized in that the second information includes the trained model and parameters used in the image processing. [Composition 7] The image processing apparatus according to configuration 5 or 6, characterized in that the aforementioned case information further includes deformation information obtained by deformation detection, the type of structure to be inspected, and its components. [Structure 8] The image processing apparatus according to any one of configurations 5 to 7, further characterized in that the aforementioned case information includes the scope of disclosure of the aforementioned case information and the appropriateness of the aforementioned case information. [Composition 9] The image processing apparatus according to configuration 6, characterized in that the presentation means selectively presents the trained model. [Configuration 10] The image processing apparatus according to configuration 9, wherein the presentation means further presents an image in which deformation detection has been performed using the trained model and an explanation of the trained model. [Composition 11] The image processing apparatus according to configuration 9 or 10, characterized in that the presentation means presents statistics on the suitability of the image for deformation detection using a selected trained model and the type of structure to be inspected. [Composition 12] The image processing apparatus according to any one of configurations 9 to 11, characterized in that the presentation means presents the parameters of a selected trained model in a configurable manner. [Composition 13] The image processing apparatus according to any one of configurations 1 to 12, further comprising detection means for performing the deformation detection based on the first information. [Composition 14] The image processing apparatus according to configuration 13, further comprising a registration means for registering the results of the deformation detection performed by the detection means as case information. [Composition 15] The image processing apparatus according to configuration 14, characterized in that the registration means can register the scope of disclosure of the case information and the appropriateness of the case information. [Composition 16] A control method for an image processing device that manages the results of anomaly detection performed on an image of an object to be inspected as case information, A step of acquiring the case information based on first information including an image used for detecting the deformation, A control method characterized by comprising the step of presenting case information obtained based on the first information. [Composition 17] A program for causing a computer to function as one of the means of an image processing apparatus described in any one of configurations 1 to 15.

[0061] The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, claims are attached to disclose the scope of the invention. [Explanation of symbols]

[0062] 100...Image processing device, 101...Control unit, 201...Acquisition unit, 202...Management unit, 203...Acquisition unit, 204...Presentation unit, 205...Detection unit, 206...Registration unit

Claims

1. In an image processing device that manages the results of abnormality detection performed on an image of an inspection object as case information, An acquisition means for acquiring the case information according to first information including an image for performing the deformation detection; and a presentation unit that presents the case information acquired by the acquisition unit.

2. 2. The image processing apparatus according to claim 1, further comprising a receiving unit for receiving the first information.

3. The image processing device according to claim 2 , wherein the first information includes information on the image for which the abnormality detection is performed and information on the structure to be inspected.

4. 4. The image processing apparatus according to claim 3, wherein the information about the structure to be inspected includes the type and material of the structure to be inspected.

5. The image processing device according to claim 1, wherein the case information includes an image on which the abnormality detection was performed and second information used in image processing for the abnormality detection.

6. The image processing device according to claim 5 , wherein the second information includes a trained model and parameters used in the image processing.

7. The image processing device according to claim 6, wherein the case information further includes deformation information obtained by the deformation detection, and the type and component of the structure to be inspected.

8. 8. The image processing apparatus according to claim 7, wherein the case information further includes a disclosure range of the case information and a suitability of the case information.

9. The image processing device according to claim 6 , wherein the presenting means presents the trained models in a selectable manner.

10. The image processing device according to claim 9, characterized in that the presentation means further presents an image on which anomaly detection has been performed using the trained model and an explanation of the trained model.

11. The image processing device described in claim 10, characterized in that the presentation means presents statistics on the suitability of the image for which abnormality detection was performed using the selected trained model and the type of structure being inspected.

12. The image processing device according to claim 11 , wherein the presenting means presents parameters of the selected trained model in a configurable manner.

13. 2. The image processing apparatus according to claim 1, further comprising a detection unit for detecting the abnormality based on the first information.

14. 14. The image processing apparatus according to claim 13, further comprising a registration unit that registers the result of the abnormality detection performed by the detection unit as case information.

15. 15. The image processing apparatus according to claim 14, wherein the registration means is capable of registering a disclosure range of the case information and a suitability level of the case information.

16. A control method for an image processing device that manages the results of detecting abnormalities in an image of an inspection object as case information, A step of acquiring the case information based on first information including an image for performing the deformation detection; and presenting case information acquired based on the first information.

17. A program for causing a computer to function as each of the means of the image processing device according to any one of claims 1 to 15.

Images