Information processing device, information processing method, and program

Abstract

To achieve a technique that can select a change to be preferentially detected for every image to be processed.SOLUTION: An information processing apparatus has: acquisition means that acquires an image to be processed; setting means that sets at least one change to the image to be processed as a detection object; selection means that selects a preferential detection object to be preferentially detected from among the detection objects; display control means that displays information indicating that one of the detection objects is selected as the preferential detection object; and detection processing means that executes, on the image to be processed, change detection processing for the detection object selected as the preferential detection object.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a technique for detecting abnormalities from an image obtained by photographing an inspection target.

Background Art

[0002] Patent Document 1 describes a method of performing image recognition processing or image analysis processing on an image obtained by photographing an inspection target such as the wall surface of a concrete structure, and detecting abnormalities such as cracks.

Prior Art Document

Patent Document

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Image recognition processing or image analysis processing of an image obtained by photographing an inspection target (image to be processed) incurs more processing costs as the types of abnormalities increase. Therefore, it becomes extremely costly to execute abnormality detection processing for all images to be processed. Thus, if it is possible to select abnormalities to be preferentially detected for each image to be processed, it is considered useful as a means for reducing processing costs.

[0005] The present invention has been made in view of the above problems, and an object thereof is to realize a technique capable of selecting abnormalities to be preferentially detected for each image to be processed.

Means for Solving the Problems

[0006] In order to solve the above problems and achieve the object, the information processing apparatus of the present invention includes acquisition means for acquiring an image to be processed Corresponding attribute information and setting means for setting at least one abnormality as a detection target for the image to be processed. A storage means for storing inspection history information regarding the inspection date and time, structure, type of deformation and quantity of deformation of the subject of inspection, and based on the structure of the subject of inspection A display means that displays the detection target set by the setting means, and based on attribute information corresponding to the image of the processing target and inspection history information stored in the storage means ​Selection means for selecting a priority detection target to be detected from among the aforementioned detection targets, and Displayed on the display means Detection target against This indicates that it has been selected as the priority detection target. prescribed It comprises a display control means for displaying information, and a detection processing means for executing deformation detection processing on the image to be processed, which is selected as the priority detection target. The display control means displays the detection target selected as the priority detection target in a changeable manner, and also displays multiple inspection history information related to the detection target. The priority detection target can be changed by selecting a detection target or inspection history information related to the detection target displayed on the display means. . [Effects of the Invention]

[0007] According to the present invention, it becomes possible to select which deformations to prioritize for detection for each image to be processed. [Brief explanation of the drawing]

[0008] [Figure 1] A schematic diagram illustrating an embodiment of the present invention. [Figure 2] Block diagram (a) and functional block diagram (b) showing the hardware configuration of the information processing device of Embodiment 1. [Figure 3] A schematic diagram illustrating Embodiment 1. [Figure 4] A flowchart illustrating the control process of Embodiment 1. [Figure 5] A diagram illustrating the method for acquiring the data to be processed. [Figure 6] A diagram explaining how to configure the detection target. [Figure 7] A diagram illustrating the method for selecting priority detection targets. [Figure 8] A diagram illustrating the method for selecting priority detection targets. [Figure 9] A diagram illustrating the screen for selecting priority detection targets. [Figure 10] A diagram illustrating the screen for selecting priority detection targets. [Figure 11] Functional block diagram of the information processing device of Embodiment 2. [Figure 12] A flowchart illustrating the control process of Embodiment 2. [Figure 13] A diagram illustrating a screen that displays the selection information and cost of the target to be detected.

Best Mode for Carrying Out the Invention

[0009] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the invention according to the claims. Although a plurality of features are described in the embodiments, not all of these plurality of features are essential for the invention, and the plurality of features may be arbitrarily combined. Further, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant explanations are omitted.

[0010] [Embodiment 1] Hereinafter, an embodiment in which the information processing apparatus of the present invention is applied to a computer apparatus used for inspecting infrastructure such as concrete structures will be described.

[0011] In Embodiment 1, an example will be described in which a computer apparatus operates as an information processing apparatus and is configured to be able to select a deformation to be preferentially detected for each image to be processed in consideration of the cost of the deformation detection process when detecting a deformation from an image obtained by imaging an inspection target.

[0012] In this embodiment, the “inspection target” is a concrete structure such as an expressway for automobiles, a bridge, a tunnel, or a dam that is an inspection target. The information processing apparatus performs a deformation detection process for detecting the presence or absence and state of deformations such as cracks using an image captured by the user of the inspection target. Further, the “deformation” includes, for example, in the case of a concrete structure, cracks, peeling, and flaking of the concrete. In addition, it includes efflorescence (whitening phenomenon), rebar exposure, rust, water leakage, dripping water, corrosion, damage (defect), cold joint, precipitate, junk, and the like.

[0013] [Overview Explanation] First, the overview of the deformation detection process of this embodiment will be described with reference to FIG. 1.

[0014] On the wall surface of a concrete structure, various types of deformations such as cracks occur due to aging deterioration and the impact of earthquakes.

[0015] Figure 1(a) shows an example of an object to be inspected, image 101 of the wall surface of a bridge. The image 101 of the object to be inspected (image to be processed) shows cracks 102, corrosion 103, and efflorescence 104. In the deformation detection process, deformation data is created from the image to be processed 101, which contains data such as the location and shape of the deformations, and is recorded in association with the image to be processed 101. One method for creating deformation data is to obtain deformation data by detecting each deformation from the image using a trained model created by machine learning or deep learning of AI (artificial intelligence).

[0016] The process of detecting deformations from images using a pre-trained model becomes more costly as the number of deformation types increases. In particular, for infrastructure structures, large-sized images captured at high resolution are used to detect minute cracks and other defects occurring in structures measuring tens of meters in size. Therefore, the cost of creating data for a large number of deformation types becomes extremely high.

[0017] Therefore, the information processing device of this embodiment selects the type of deformation to be detected according to the image to be processed and executes the deformation detection process. The information processing device of this embodiment selects the initial value of the priority detection target deformation to be detected first from among the deformations set as detection targets. One method for selecting the priority detection target is to use inspection history information. Figure 1(b) is image 111, which was taken in the past (for example, 5 years ago) of the same bridge and the same area as image 101. Figure 1(c) shows an example of deformation data 121 detected from image 111. The deformation data 121 includes three types of deformation data: crack data 122, corrosion data 123, and efflorescence data 124. In this way, the types of deformation (cracks, corrosion, efflorescence) recorded in past inspections are selected as priority detection targets. Figure 1(d) shows an example of the priority detection target selection screen 131. On the selection screen 131, three deformations are selected as priority detection targets 133 from among the detection targets 132. As described above, the information processing device of this embodiment can select an initial value for a priority detection target 133 from among the detection targets 132, according to the inspection history information of the image to be processed. The initial value for the priority detection target can be arbitrarily changed by the user. When the user operates the detection execution button 134 on the selection screen 131, the deformation detection process is executed. Figure 1(e) shows an example of the deformation detection process result 141. The deformation detection process result 141 includes crack data 142, corrosion data 143, and efflorescence data 144. In this way, by displaying the deformation to be detected according to the image to be processed in a selectable manner, it becomes possible to execute deformation detection processing that takes cost into consideration.

[0018] <Hardware Configuration> Next, the hardware configuration of the information processing device of Embodiment 1 will be described with reference to Figure 2(a).

[0019] Figure 2(a) is a block diagram showing the hardware configuration of the information processing device 200 of Embodiment 1.

[0020] In Embodiment 1, a computer device operates as an information processing device 200. The processing of the information processing device in this embodiment may be implemented by a single computer device, or the functions may be distributed among multiple computer devices as needed. The multiple computer devices are connected to each other in a way that allows them to communicate with one another.

[0021] The information processing device 200 includes a control unit 201, a non-volatile memory 202, a work memory 203, a storage device 204, an input device 205, an output device 206, a network interface 207, and a system bus 208.

[0022] The control unit 201 includes a CPU, MPU, and other arithmetic processing processors that comprehensively control the entire information processing device 200. The non-volatile memory 202 is a ROM that stores programs and parameters executed by the processor of the control unit 201. Here, the program refers to a program for executing the processes of Embodiments 1 and 2, which will be described later. The work memory 203 is a RAM that temporarily stores programs and data supplied from external devices, etc. The work memory 203 holds data obtained by executing the control process shown in Figure 4, which will be described later.

[0023] The storage device 204 is an internal device such as a hard disk or memory card built into the information processing device 200, or an external device such as a hard disk or memory card detachably connected to the information processing device 200. The storage device 204 includes memory cards and hard disks made of semiconductor memory or magnetic disks. The storage device 204 also includes a storage medium consisting of a disk drive that reads / writes data to / from optical discs such as DVDs and Blu-ray Discs.

[0024] The input device 205 is an operating component such as a mouse, keyboard, or touch panel that accepts user input and outputs operation instructions to the control unit 201. The output device 206 is a display device such as a display or monitor made of an LCD or organic EL, which displays data held by the information processing device 200 or data supplied from external devices. The network interface 207 is connected to a network such as the Internet or a LAN (Local Area Network) for communication. The system bus 208 includes an address bus, a data bus, and a control bus that connect the various components 201 to 207 of the information processing device 200 for data exchange.

[0025] The non-volatile memory 202 stores the operating system (OS), which is the basic software executed by the control unit 201, and applications that work in cooperation with the OS to realize advanced functions. In this embodiment, the non-volatile memory 202 also stores applications that enable the control processing described later by the information processing device 200.

[0026] The control processing of the information processing device 200 in this embodiment is realized 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 information processing device 200. The OS of the information processing device 200 may also have software for realizing the control processing in this embodiment.

[0027] <Functional Configuration> Next, with reference to Figure 2(b), the functional blocks of the information processing device 200 of Embodiment 1 will be described.

[0028] Figure 2(b) is a functional block diagram of the information processing device 200 of Embodiment 1.

[0029] The information processing device 200 includes a storage unit 221, a management unit 222, a data acquisition unit 223, a detection target setting unit 224, a priority detection target selection unit 225, a display control unit 226, and a detection processing unit 227. Each function of the information processing device 200 is configured by hardware and / or software. Each function unit may be composed of one or more computer devices or server devices and configured as a system connected by a network. Furthermore, if each function unit shown in Figure 2(b) is implemented by hardware instead of software, it is sufficient to have the circuit configuration corresponding to each function unit in Figure 2(b).

[0030] The management unit 222 manages the registration, deletion, and updating of image data and inspection history information to be processed, which are stored in the storage unit 221.

[0031] The data acquisition unit 223 acquires the image data to be processed.

[0032] The detection target setting unit 224 sets at least one abnormality as a detection target based on the inspection history information stored in the storage unit 221.

[0033] The priority detection target selection unit 225 selects the priority detection target deformations from among the detection target deformations to be detected first.

[0034] The display control unit 226 creates display data such as the image to be processed, the object to be detected, the selection information of the object to be detected, and the inspection results, and outputs it to the output device 206. The output device 206 then displays the image based on the display data on the display unit.

[0035] The detection processing unit 227 executes deformation detection processing on the image data to be processed, selecting the detection target as the priority detection target.

[0036] Details of the processing and functions of the data acquisition unit 223, detection target setting unit 224, priority detection target selection unit 225, display control unit 226, and detection processing unit 227 will be described later in Figure 4.

[0037] <Explanation of Inspection History Information> Next, with reference to Figure 3, the inspection history information used in the control processing of the information processing device 200 of this embodiment will be explained.

[0038] The inspection history information in this embodiment includes attribute information about the structure being inspected, inspection results for each component of the structure being inspected, image data to be processed, and attribute information about the deformation, and these are managed in association with each other.

[0039] Figure 3 shows an example of inspection history information for an infrastructure structure in this embodiment.

[0040] The structure table 301 in Figure 3(a) is a table that holds attribute information about the structure to be inspected. In the structure table 301, the structure ID is a unique identifier that identifies the structure information, and each structure ID has the following items: structure name, structure type, installation location, and completion date. In addition, the structure table 301 may also hold other attribute information, such as the total length and height indicating the size of the structure, and the distance from the coast and elevation related to the installation location. Furthermore, additional attribute information may be held for each structure type.

[0041] The member inspection results table 311 in Figure 3(b) is a table that holds inspection results for each member of a structure. In the member inspection results table 311, the member inspection ID is identification information that identifies the inspection result, and each member inspection ID has the following items: inspection date, member name, member type, structure ID, member damage level, and member soundness level. The member inspection results table 311 is associated with the structure table 301 by the structure ID. The member inspection results table 311 may also hold other information related to the inspection results as items.

[0042] The image table 321 in Figure 3(c) is a table for managing image data of the wall surface of a structure. The image table 321 is associated with the member inspection result table 311 by member inspection ID. It is desirable that the image data of the wall surface of a structure be managed in association with drawings (e.g., structural development drawings). Therefore, the image table 321 holds coordinate values ​​that indicate the positional relationship of each image data on the drawing. For example, the image data with image ID IMG001 has (Xul001,Yul001) and (Xbr001,Ybr001) as the top-left and bottom-right vertex coordinates of the drawing, respectively. By holding the vertex coordinates of the drawing for each image to be processed in this way, the correspondence between deformation data and image data, which will be described later, can be determined. Note that the image table 321 may also hold other attribute information related to the image data.

[0043] The deformation table 331 in Figure 3(d) is a table that manages deformation data related to the position and shape of deformations on the wall surface of a structure, which are detected using image data taken of the wall surface of the structure. The coordinate column and numerical column in the deformation table 331 are the coordinate values ​​that make up the deformation data and the attribute value that represents the width of the deformation at those coordinates. For example, crack C001 is represented by a continuous pixel consisting of n points from (Xc001_1, Yc001_1) to (Xc001_n, Yc001_n). In this embodiment, the deformation data is assumed to be represented by pixels, but it may also be represented by vector data such as polylines or curves composed of multiple points. When deformation data is represented by vector data, the data size is reduced and the representation becomes simpler.

[0044] The information processing device 200 of this embodiment can use the inspection history information to obtain the type and quantity of deformations detected from the image to be processed, as well as the time-series progression of deformation occurrence.

[0045] <Control Processing> Next, referring to Figure 4, the control processing of the information processing device 200 in this embodiment is shown as a flowchart.

[0046] The process in FIG. 4 is realized by the control unit 201 of the information processing apparatus 200 shown in FIG. 2(a) expanding and executing the program stored in the non-volatile memory 202 in the work memory 203, controlling each component shown in FIG. 2(a), and operating as each functional unit shown in FIG. 2(b). Further, the process in FIG. 4 is started when the information processing apparatus 200 receives an instruction to start the deformation detection process from the input device 205.

[0047] <S401: Acquisition Process of Processing Target Data> The data acquisition unit 223 performs a process of acquiring the image data to be processed. In this embodiment, the image data obtained by photographing the wall surface of a concrete structure is acquired as the processing target data. Examples of the method for acquiring the processing target data include a method of displaying a screen for receiving a user operation and acquiring the processing target data according to the user's operation instruction. The acquisition process of the processing target data according to the user's operation instruction will be described using FIG. 5.

[0048] FIG. 5(a) illustrates a screen 501 for receiving the input of the processing target data. After receiving the input of the image on the screen 501, the data acquisition unit 223 acquires the processing target data according to the operation of the input completion button 502. FIG. 5(b) illustrates the image 511 input on the screen 501. The image 511 is an image obtained by photographing the wall surface of a concrete structure and includes cracks 512, corrosion 513, and efflorescence 514. Note that "efflo" in the figure is an abbreviation for efflorescence.

[0049] In the acquisition process of the data to be processed, additional acquisition of attribute information regarding the photographed structure may be performed. FIG. 5(c) illustrates a screen 521 for receiving the input of attribute information and an image regarding the data to be processed. After receiving the input on the screen 521, in response to the operation of the input completion button 525, the data acquisition unit 223 acquires the acquisition date 522 of the image, the name 523 of the structure to be photographed, and the member type 524 of the object to be photographed as the attribute information regarding the data to be processed and the image. The options for the structure name 523 and the member type 524 can be stored in the storage unit 221 in advance, and the management unit 222 can acquire them from the storage unit 221 and display them on the screen 521. When acquiring attribute information not stored in the storage unit 221, a method of displaying a screen for receiving the input of various attribute information, such as the screen 531 illustrated in FIG. 5(d), can be used. In the data acquisition process, the image recognition result or the image analysis result of the data to be processed may be acquired as attribute information. For example, the average pixel value, the variance value, etc. can be calculated by the image recognition process or the image analysis process and acquired as the attribute information regarding the data to be processed.

[0050] <S402: Detection Target Setting Process> The detection target setting unit 224 performs a process of setting the type of change of the detection target. As a method of setting the detection target, for example, there is a method in which the management unit 222 acquires the change type information stored in the storage unit 221 in advance and sets it as the detection target. FIG. 6(a) illustrates the detection target 601.

[0051] In the detection target setting process, the detection target may be switched according to the image to be processed. For example, a list of different deformation types for each type of structure is stored in the storage unit 221. The detection target setting unit 224 acquires the list of deformation types according to the structure type of the attribute information regarding the image to be processed and sets it as the detection target. FIG. 6(b) illustrates the detection targets for each type of structure. The detection targets 611 and 612 illustrate the detection targets for bridges and tunnels respectively. Thus, it is desirable to switch the detection target based on the attribute information of the image to be processed. When switching the detection target according to the attribute information of the image to be processed, it is necessary to acquire the attribute information of the image to be processed in S401.

[0052] <S403: Selection Process of Priority Detection Target> The priority detection target selection unit 225 performs a process of selecting the deformation type to be preferentially detected as the priority detection target. As a method for selecting the priority detection target, for example, a method using the inspection history information of the structure to be inspected can be cited. The selection process of the priority detection target will be described using FIGS. 7 and 8.

[0053] FIG. 7(a) illustrates the attribute information 701 regarding the image to be processed. The attribute information 701 can be acquired together with the image to be processed in S401. FIG. 7(b) illustrates the inspection history information 711 showing the inspection results for each member of the structure. The inspection history information 711 holds, as items, the deformation type occurring in the member and the quantity for each deformation type. When selecting the deformation type of the priority detection target, first, a row in the inspection history information 711 where the structure name and member type match the attribute information 701 is searched. Next, among the search results, the row with the inspection date closest to the acquisition date of the attribute information 701 is narrowed down. FIG. 7(c) illustrates the inspection history information 721 obtained by the narrowing down, and the deformation type of the inspection history information 721 is selected as the priority detection target. FIG. 7(d) illustrates the deformation type 731 selected as the priority detection target. Thus, the deformation type that occurred in the previous inspection can be acquired from the inspection history information and selected as the priority detection target.

[0054] In the process of selecting priority detection targets, the inspection history information used may be changed depending on the image to be processed. For example, all types of deformation of the entire structure related to the image to be processed may be selected as priority detection targets. Alternatively, among the deformations that occurred within the same structure name and within the same member as the attribute information of the image to be processed, only past deformations that occurred within the shooting range may be selected as priority detection targets. The position of deformation data detected in past inspections on the drawing is stored in the storage unit 221 as coordinate information as shown in Figure 3(d). In addition, the position of the image to be processed on the drawing can be additionally acquired by the data acquisition unit 223 along with the image to be processed as attribute information related to the image. In this way, since the positions of deformations that occurred in past inspections on the drawing and the positions of the image to be processed on the drawing are obtained, only deformations that occurred within the shooting range of the image to be processed can be selected as priority detection targets.

[0055] In the process of selecting priority detection targets using inspection history information, inspection history information of a structure different from the structure being inspected may be used. For example, suppose the attribute information and inspection history information for the image to be processed are shown in Figures 8(a) and 8(b), respectively. The inspection history information 811 in Figure 8(b) does not contain inspection history information identical to the structure name in the attribute information 801 of Figure 8(a). In such cases, inspection history information of a structure similar to the structure being inspected can be used. Specifically, the system searches for rows in the inspection history information 811 where the structure type and member type match those in the attribute information 801. From the search results, the system narrows down the results to rows where the number of years used at the time of inspection is closest to the number of years used in the attribute information 801. Figure 8(c) shows an example of the inspection history information 821 obtained through this narrowing down process. The deformation type in the inspection history information 821 is then selected as the initial value for the priority detection target. Figure 8(d) shows an example of the priority detection target 831. In this way, priority detection targets can be selected based on the inspection history information of a structure similar to the structure being inspected. While the example described uses matching structural type and component type as a condition for determining structures highly similar to the structure being inspected, other attribute information may be used. For example, information such as the total length and location of the structure may be used. Alternatively, multiple items may be used, and the structure with the most matching items may be considered the most similar structure. Furthermore, when determining similar structures, the degree of similarity may be determined based on the occurrence of deformation on the structural wall surface. Specifically, the type of deformation and the quantity for each deformation type may be calculated from the inspection history information over time, and the degree of similarity may be calculated based on the progression of deformation occurrence to determine the most similar structure.

[0056] The selection process of the priority detection target is not limited to the method described above, and other methods may be used. For example, the deformation types to be the priority detection targets may be stored in the storage unit 221 in advance, and the priority detection target selection unit 225 may obtain and select them from the storage unit 221. In the inspection of special infrastructure, there may be cases where it is desired to check the presence or absence of specific deformations each time. In such cases, it is desirable to register fixed deformation types as the priority detection targets in advance. For the method of registering the priority detection targets in advance, specific deformation types (e.g., cracks, efflorescence) may be registered, or a method of accepting input from the user may be used. Fig. 8(e) is a screen for accepting input of the priority detection targets provided by the application implementing the processing of this embodiment, and illustrates a screen 841 for registering the initial values of the priority detection targets for each user. On the screen 841, when the user operates the OK button 843 with any of the detection targets 842 in a selected state, the priority detection target selection unit 225 registers the selected detection target 842 as the initial value of the priority detection target for the user 844. Thus, when using the method of accepting input from the user, it is desirable to register it in association with the properties related to the application implementing the processing of this embodiment. Note that other properties may be used as the property for registering the initial values of the priority detection targets. For example, the initial values of the priority detection targets may be registered for a user group that manages a plurality of users, or for an image group that manages images in the application.

[0057] <S404: Display Process of Selection Information of Detection Target> The display control unit 226 creates and displays display data to show selection information indicating that the detection target has been selected as a priority detection target. In this embodiment, the selection information for the detection target is created based on the priority detection target selected in S403. Specifically, the display control unit 226 selects one deformation type from the detection targets set in S402 and determines whether or not it is included in the priority detection target selected in S403. If the display control unit 226 determines that the deformation type of the selected detection target is included in the priority detection target, it determines that the deformation type has already been selected as a priority detection target. This determination process is repeated for all deformation types of detection targets to set the initial value of the selection information for the detection target. Figure 9(a) illustrates a screen 901 that displays the selection information for the detection target 902. In the detection target 902, the selected deformation type 903 corresponds to the deformation type of the detection target selected as a priority detection target. Note that the image 904 of the processing target is displayed on screen 901 along with the selection information for the detection target 902. Displaying the image to be processed in this way makes it easier for the user to determine whether the selection of the priority detection target is appropriate (i.e., whether the detection target is selected according to the image being processed).

[0058] It is desirable that the screen displayed in S404 has a function to accept changes to the selection information of the detection target. For example, if the user wants to add "float" as a detection target on screen 901, they can update the selection information by checking checkbox 905. It is also desirable that the display data created by the display control unit 226 has a reset function to return the selection information of the detection target to its initial value. For example, on screen 901, after changing the selection information of the detection target, checking checkbox 906 will return the selection information of detection target 902 to the initial value of the priority detection target selected in S403.

[0059] Furthermore, it is desirable that the screen displayed in S404 displays inspection history information. Figure 9(b) illustrates a screen 911 that displays the detection target 912, the image 913 to be processed, and past inspection results 914 for the same shooting range as the image 913 to be processed. Past inspection results 914 can be created, for example, by superimposing images and deformation data taken in the previous inspection (e.g., 5 years ago). By displaying inspection history information in this way, it becomes easier for the user to determine whether the selection of priority detection targets is appropriate. In addition, time-series data calculated based on the inspection history information may be displayed as the inspection history information. Figure 9(c) illustrates a screen 921 that displays the selection information of the detection target 922 and a graph 923 that aggregates the occurrence amount of each deformation for each inspection period as time-series data. By displaying time-series data related to inspection history information in this way, it becomes easier for the user to confirm the increasing trend of deformation due to aging.

[0060] In addition, if the selection of the priority detection target in S404 is inappropriate, the user can reselect the type of change in the detection target. In such a case, it is desirable to provide a function for collectively switching the selection information of the detection target. For example, a method of switching the selection information of the detection target is to display a plurality of inspection history information and allow the user to select from them. FIG. 10(a) illustrates a screen 1001 on which the selection information of the detection target 1002, the transition 1003 of the occurrence of a change in the inspected structure, and the transitions 1004 and 1005 of the occurrence of a change in similar structures different from the inspection target are displayed. Here, it is assumed that the initial value of the priority detection target selected from the detection target 1002 is selected as the priority detection target according to the transition 1004 of the occurrence of a change in the similar structure. On the screen 1001, the transition 1003 of the occurrence of a change in the inspected structure and the transition 1004 of the occurrence of a change in the similar structure are significantly different despite a high degree of coincidence of the attribute information. In such a case, it is desirable to switch the detection target by displaying the transitions of the occurrence of changes in a plurality of structures simultaneously and allowing the user to select. For example, on the screen 1001, when the user selects a transition 1005 of the occurrence of a change close to the transition 1003 of the occurrence of a change in the inspected structure, the display control unit 226 receives the user's selection operation, and the priority detection target selection unit 225 performs the selection process of the priority detection target based on the user's selection instruction. Then, the display control unit 226 creates and displays display data in which the selection information of the detection target is changed according to the reselected priority detection target (screen 1011 in FIG. 10(b)). Thus, it is desirable to collectively switch the selection information of the detection target.

[0061] <S405: Change Detection Process> The detection processing unit 227 executes deformation detection processing on the image data to be processed, prioritizing the detection of abnormalities. One way to start the execution of the deformation detection processing is, for example, based on user instructions. For example, on screen 901 in Figure 9(a), which displays the selection information 903 of the detection target 905, when the user operates the detection execution button 907, the detection processing unit 227 executes detection processing on the image data to be processed for the deformation type 903 selected on screen 901. For example, a trained model learned by machine learning can be used to execute the detection processing. When executing detection processing using a trained model, it is necessary to select the trained model and specify a number of parameters. The method for specifying the trained model and parameters may be to use a predetermined trained model and parameters for each deformation type, or the trained model and parameters may be switched and specified according to the image characteristics of the image to be processed. Alternatively, the user may specify the trained model and parameters when executing the detection processing.

[0062] Then, after executing the detection process in S405, the display control unit 226 displays display data indicating the detection process result and terminates the process shown in Figure 4.

[0063] [Modified example of Embodiment 1] In the embodiment 1 described above, an example was explained in which, according to the image data to be processed, the priority detection target deformations to be detected are selected and displayed from among the detection target deformations, and then the deformation detection process is executed according to the user's instructions. When performing deformation detection processing on a large number of images, it is very time-consuming to check the selection information of the detection targets for each image to be processed. In such cases, the detection process may be executed without user instructions. Specifically, in S401, the image data to be processed is acquired, and the detection target setting process and the priority detection target selection process are performed. After that, the selection information of the detection targets is displayed, and the detection process is started automatically. After the deformation detection process has started, the user can check the progress of the deformation detection process and the priority detection target deformations for each image to be processed. In this way, by executing the deformation detection process without waiting for user instructions, the user's operations related to the deformation detection process can be made more efficient.

[0064] Furthermore, in the above-described embodiment 1, an example was explained in which deformations to be detected with priority are selected according to the image to be processed and the detection process is executed. When detecting many types of deformations, the processing time increases, and the user has to wait until they can view the detection results. In such cases, it is desirable to set a priority for the deformations to be detected. For example, among the deformations to be detected, deformations identical to the priority detection target are given priority for detection processing, and the detection results of deformations identical to the priority detection target are displayed so that the user can check them. Then, while the user is checking, the detection process for the remaining deformations is performed. In this way, by setting a priority for the deformations to be detected, the user can check whether there are any deformations that they want to check with priority.

[0065] According to Embodiment 1 described above, the deformations to be detected as priority are selected and displayed from among the deformations to be detected according to the image to be processed. This allows the user to consider the cost required for deformation detection processing and other factors, and then confirm whether the deformations selected as priority detection targets are appropriate before executing the deformation detection processing.

[0066] [Embodiment 2] In Embodiment 1, an example was described in which the deformation to be detected is selected and displayed according to the image to be processed, and the detection process is executed. However, it is desirable that the user can simultaneously check the cost required for the deformation detection process when confirming the selection information of the detection target. An example of cost in Embodiment 2 is the cost (amount) required for the deformation detection process. In Embodiment 2, by displaying the selection information of the detection target and the cost required for the deformation detection process, the user can more easily check and change the deformation to be detected while considering the cost.

[0067] Hereinafter, the description will focus on the parts different from those in Embodiment 1. In Embodiment 2, an example using a metered billing type paid application according to the detection processing amount will be described. However, the applications and services implementing the processing of Embodiment 2 may use other forms. For example, it can also be implemented in the form of a cloud type service such as SaaS (Software as a Service).

[0068] The hardware configuration of the information processing apparatus 200 in Embodiment 2 is the same as that in Fig. 2(a).

[0069] Fig. 11 is a functional block diagram of the information processing apparatus 200 in Embodiment 2. The functional configuration in Fig. 11 is different from the functional configuration shown in Fig. 2(b) of Embodiment 1 in that a calculation unit 1101 is added. The calculation unit 1101 is a functional unit of the control unit 201 and performs a process of calculating the cost related to the detection process. In Embodiment 2, the display control unit 226 creates display data for displaying the selection information of the detection target and the cost calculated by the calculation unit 1101.

[0070] Fig. 12 is a flowchart showing the processing of the information processing apparatus 200 in Embodiment 2.

[0071] In Fig. 12, the same steps as those in Fig. 4 of Embodiment 1 are given the same step numbers.

[0072] After selecting the priority detection target in S403, the process proceeds to the process of S1201.

[0073] <S1201: Calculation of cost related to detection processing> In S120, the calculation unit 1101 performs a process of calculating the cost related to the change detection process. In Embodiment 2, as the cost required for the change detection process, the amount estimated for executing the detection process is calculated as the estimated amount. Equation 1 for calculating the estimated amount Cp is shown below. (Equation 1) In Equation 1 of TIFF0007842615000001.tif34144, parameter I indicates the number of target data acquired in S401, i.e., the number of images to be processed. Parameter Pn indicates the unit price per image for the nth deformation type out of the N deformation types to be detected. A basic method for setting parameter Pn is to pre-store a uniform unit price (e.g., 1,000 yen) for all deformation types in the storage unit 221, and have the management unit 222 retrieve it from the storage unit 221. Different unit prices may be set for each deformation type. Since the computational resources required for deformation detection processing differ for each deformation type, it is desirable to change the unit price for each deformation type. Parameter N is the total number of deformation types to be detected, and can be obtained from the selection information of the detection targets. In this way, the estimated cost required for deformation detection processing can be obtained using Equation 1.

[0074] In S1201, costs other than the estimated amount may be calculated. For example, the predicted time, which is the time required to execute the detection process, may be used. Equation 2 for calculating the predicted time Ct is shown below. (Formula 2) In Equation 2 of TIFF0007842615000002.tif39151, parameter I, as in Equation 1, indicates the number of target data acquired in S401. Parameter Tn indicates the processing time required for the deformation detection process per image for the nth deformation type out of the N deformations to be detected. One method for setting parameter Tn is to pre-store a uniform processing time (e.g., 10 minutes) for all deformation types in the storage unit 221, and then have the management unit 222 retrieve it from the storage unit 221. Different processing times may be set for each deformation type. Since the time required for deformation detection processing differs for each deformation type, it is desirable to set a processing time for each deformation type. Also, parameter N, as in Equation 1, is the total number of deformation types to be detected and can be obtained from the selection information of the detection targets. In this way, the predicted time can be calculated using Equation 2.

[0075] Equations 1 and 2 are equations for calculating the cost proportional to the number of images to be processed, but they may also be equations for calculating the cost considering the image size. For example, in Equation 1 for calculating the estimated amount Cp, the parameter I is read as the total area of the images to be processed, and the parameter Pn is read as the unit price per unit area in the nth deformation type. The setting method of the parameter Pn can be set in the same way as in the case of Equation 1. Thus, the estimated amount can be calculated when considering the image size. Similarly, in the case of Equation 2 for calculating the prediction time Ct, the parameter I is read as the total area of the images to be processed, and the parameter Tn is read as the processing time per unit area in the nth deformation type. The setting method of the parameter Tn can be set in the same way as in the case of Equation 2.

[0076] As the cost calculation process in the calculation unit 1101, a plurality of costs may be calculated. That is, both the estimated amount Ct and the prediction time Ct are calculated using Equation 1 and Equation 2. The cost calculated by the calculation unit 1101 is used in the process of S1202 described later.

[0077] <S1202: Display Process of Selection Information of Detection Target and Cost> In S1202, the display control unit 226 performs a process of creating and displaying display data for displaying the selection information of the detection target and the cost required for the deformation detection process.

[0078] Figure 13(a) illustrates screen 1301 displayed in S1202. Screen 1301 displays the detection target 1302, the number of images to be processed 1303, and the estimated cost 1304. By displaying the selection information for the detection target 1302 and the estimated cost 1304, as shown in screen 1301, the user can easily check and change the deformation of the detection target while considering the cost. The initial value of the priority detection target selected from the detection target 1302 is selected according to the priority detection target, similar to the process in S404 in Figure 4. The estimated cost 1304 can be calculated using the process in S1201 described above. Specifically, in Equation 1, it can be calculated by substituting the value of the number of images 1303 (100) for parameter I, the number of selected deformation types (2) for parameter N, and 1,000 for all deformation types for the unit price Pn for each deformation type, as shown in unit price 1305.

[0079] It is desirable that the screen displayed in S1202 shows statistical data indicating the occurrence status of each type of deformation. Figure 13(b) illustrates a screen 1311 that displays the detection target 1312, multiple costs 1313 (estimated amount, predicted time), and statistical data 1314 for each type of deformation. The statistical data 1314 includes the number of past deformation occurrences, total length (area), and the trend of the number of deformation occurrences for the same component as the inspection target. By displaying the costs 1313 and statistical data 1314 in this way, it becomes easier for the user to determine whether the selection information for the detection target 1312 is commensurate with the cost. The statistical data 1314 can be calculated based on the inspection history information shown in Figure 3. In addition, the statistical data 1314 may also display other data, such as the number of occurrences per unit area (density) for each type of deformation, or the amount of deformation that occurred in the inspection before last. Thus, it is desirable to display statistical data for each detection target based on inspection history information.

[0080] In S1203, the display control unit 226 determines whether the selection information of the detection target has been updated. If the display control unit 226 determines that there has been no update, it proceeds to S405; if it determines that there has been an update, it returns to S1201.

[0081] According to Embodiment 2 described above, by displaying the deformation of the priority detection target selected as the detection target and the cost required for the deformation detection process, the user can select a priority detection target from among the deformations to be detected while considering the cost, and then execute the deformation detection process after selecting an appropriate detection target.

[0082] [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.

[0083] The disclosures herein include the following information processing devices, information processing methods, and programs. [Configuration 1] An acquisition means for acquiring the image to be processed, A setting means for setting at least one anomaly as a target for detection in the image to be processed, A selection means for selecting a priority detection target to be detected from among the aforementioned detection targets, A display control means that displays information indicating that the detection target has been selected as the priority detection target, An information processing apparatus characterized by having detection processing means that performs deformation detection processing on the image to be processed, which is selected as the priority detection target. [Configuration 2] The information processing apparatus according to configuration 1, characterized in that the acquisition means acquires attribute information of the image to be processed. [Configuration 3] The information processing apparatus according to configuration 2, characterized in that the attribute information includes information about the object to be inspected and information about the deformation detected from the image of the object to be processed. [Structure 4] The information processing apparatus according to any one of configurations 1 to 3, characterized in that the selection means selects the priority detection target from among the detection targets based on inspection history information of the inspection target. [Composition 5] The information processing device according to configuration 4, characterized in that the inspection history information includes inspection history information for the same inspection target as the image to be processed. [Composition 6] The information processing apparatus according to configuration 4, characterized in that the inspection history information includes inspection history information for an inspection target different from the inspection target related to the image to be processed. [Composition 7] The information processing apparatus according to any one of configurations 4 to 6, characterized in that the display control means displays the image to be processed and the inspection history information. [Structure 8] The information processing apparatus according to any one of configurations 1 to 7, characterized in that the display control means displays the information in a modifiable manner. [Composition 9] The information processing apparatus according to any one of configurations 1 to 8, characterized in that the detection processing means performs the deformation detection process in response to user instructions or the display of the information. [Configuration 10] The information processing apparatus according to any one of configurations 1 to 9, characterized in that the detection processing means executes an abnormality detection process for a detection target selected as the priority detection target according to a predetermined priority order. [Composition 11] The information processing apparatus according to any one of configurations 1 to 10, characterized by having a calculation means for calculating the cost related to the deformation detection process of the target to be detected. [Composition 12] The information processing device according to configuration 11, characterized in that the cost is calculated to be at least one of the amount of money or processing time required to perform the abnormality detection process for the object to be detected. [Composition 13] The information processing apparatus according to configuration 11 or 12, characterized in that the display control means displays the detected target and the cost. [Composition 14] The image to be processed is an image taken of the object to be inspected. The information processing apparatus according to configuration 1, characterized in that the deformation includes cracks. [Composition 15] The acquisition means includes the step of acquiring the image to be processed, The setting means includes the step of setting at least one deformation as a target for detection in the image to be processed, The selection means includes the step of selecting a priority detection target to be detected from among the detection targets, The display control means includes the step of displaying information indicating that the detection target has been selected as the priority detection target, An information processing method characterized by comprising the step of a detection processing means performing a deformation detection process on an image to be processed that has been selected as a priority detection target. [Composition 16] A program for causing a computer to function as one of the information processing devices described in any one of the configurations 1 to 14.

[0084] 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]

[0085] 200...Information processing device, 201...Control unit, 221...Storage unit, 222...Management unit, 223...Data acquisition unit, 224...Detection target setting unit, 225...Priority detection target selection unit, 226...Display control unit, 227...Detection processing unit

Claims

1. A means for obtaining attribute information corresponding to the image to be processed, A storage means for storing inspection history information regarding the inspection date and time, structure, type of deformation, and quantity of deformation of the subject of inspection, A setting means for setting at least one deformation to be detected in the image to be processed based on the structure to be inspected, A display means for displaying the detection target set by the setting means, A selection means for selecting a priority detection target from among the detection targets based on attribute information corresponding to the image to be processed and inspection history information stored in the storage means, A display control means that displays predetermined information indicating that a detection target displayed on the display means has been selected as the priority detection target, The system includes a detection processing means that performs deformation detection processing on the image to be processed, which is selected as the priority detection target. The display control means displays the detection target selected as the priority detection target in a changeable manner, and also displays multiple inspection history information related to the detection target. The information processing device is characterized in that the priority detection target can be changed by selecting a detection target displayed on the display means or inspection history information relating to the detection target.

2. The information processing device according to claim 1, characterized in that the attribute information includes the date and time of inspection of the subject of inspection, the structure, and the members.

3. The information processing apparatus according to claim 1, characterized in that the inspection history information relating to the detection target includes inspection history information for the same inspection target as the inspection target relating to the image of the processing target.

4. The information processing apparatus according to claim 1, characterized in that the inspection history information relating to the detection target includes inspection history information of an inspection target different from the inspection target relating to the image of the processing target.

5. The information processing apparatus according to claim 1, characterized in that the detection processing means performs the deformation detection process in response to a user instruction or the display of information indicating a detection target selected as the priority detection target.

6. The information processing apparatus according to claim 1, characterized in that the detection processing means performs deformation detection processing on a detection target selected as a priority detection target according to a predetermined priority order.

7. The information processing apparatus according to claim 1, characterized in that it has a calculation means for calculating the cost related to the deformation detection process of the target to be detected.

8. The information processing apparatus according to claim 7, characterized in that the cost is calculated to be at least one of the amount of money or processing time required to perform the deformation detection process for the object to be detected.

9. The information processing apparatus according to claim 7, characterized in that the display control means displays the detected target and the cost.

10. The image to be processed is an image taken of the object to be inspected. The information processing apparatus according to claim 1, characterized in that the deformation includes cracks.

11. An information processing method performed by an information processing device, The aforementioned information processing device is It has a storage means for storing inspection history information regarding the inspection date and time, structure, type of deformation, and quantity of deformation of the subject of inspection. The aforementioned information processing method is The steps include obtaining attribute information corresponding to the image to be processed, The steps include setting at least one deformation as a target for detection in the image to be processed based on the structure to be inspected, The steps include: displaying the set detection target on the display unit; A step of selecting a priority detection target from among the detection targets based on attribute information corresponding to the image to be processed and inspection history information stored in the storage means, The steps include: displaying predetermined information on the display unit indicating that the detection target displayed on the display unit has been selected as the priority detection target; The process includes the step of performing deformation detection processing on the image to be processed, which is selected as the priority detection target. In the step of displaying the predetermined information, the detection target selected as the priority detection target is displayed in a way that allows it to be changed, and multiple inspection history information related to the detection target is displayed. The information processing method is characterized in that the priority detection target can be changed by selecting a detection target displayed on the display unit or inspection history information related to the detection target.

12. A program for causing a computer to function as each of the means of an information processing device described in any one of claims 1 to 10, excluding the storage means and the display means.

Images