Information processing device, information processing method, and program

Abstract

To solve the problem of reducing detection omission because a detection object of a damage or the like has been singly detected.SOLUTION: An information processing device includes image acquisition means for acquiring an image obtained by photographing a structure in which a detection object exists, first detection data acquisition means for acquiring first detection data being data related to a plurality of detection objects detected from the image, combination information acquisition means for acquiring combination information about a combination of a first detection object included in the plurality of detection objects and a second detection object related to the first detection object, and analysis means for analyzing the first detection object of the first detection data to generate an analysis result on the basis of the second detection object combined with the first detection object shown by the combination information.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] In the inspection of concrete structures such as bridges and tunnels, image inspection is performed using images of the wall surface of the structure. In this image inspection, image recognition technology such as machine learning is used to detect damage such as cracks on the wall surface. Since the detection result includes detection omissions where damage is not detected, the inspection worker checks the detection result. Checking the detection result is time-consuming. Patent Document 1 discloses a technology for detecting damage at each detection point from a road surface image and determining image positions with a high possibility of false detection. By using the technology of Patent Document 1, it is possible to efficiently check and correct false detection results included in the detection result. Patent Document 2 discloses a technology for detecting closed cracks. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7156527 [Patent Document 2] Patent No. 5645730 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the conventional technology, the detection target such as damage was detected individually, making it difficult to reduce detection omissions. [Means for solving the problem]

[0005] In order to solve this problem, for example, an information processing device of the present invention has the following configuration. an image acquisition means for acquiring an image of a structure in which a detection target is present; a first detection data acquisition means for acquiring first detection data which is data relating to a plurality of detection targets detected from the image; a combination information acquisition means for acquiring combination information regarding a combination of a first detection target included in the plurality of detection targets and a second detection target related to the first detection target; an analysis means for analyzing the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information, and generating an analysis result; Equipped with. Effect of the Invention

[0006] According to the present invention, the first detection data, which is the detection result of the first detection target, is analyzed based on combination information regarding the combination of the first detection target and an associated second detection target, thereby reducing the risk of missing out on detecting the first detection target compared to detecting a specific detection target alone. [Brief description of the drawings]

[0007] [Figure 1] 1A and 1B are a block diagram and a functional block diagram illustrating a hardware configuration of an information processing apparatus according to a first embodiment. [Diagram 2] FIG. 4 is a diagram for explaining combination information of detection targets. [Diagram 3] FIG. 2 is a diagram illustrating an image and detection data. [Figure 4] 4 is a flowchart illustrating a control process according to the first embodiment. [Diagram 5] FIG. 13 is a diagram illustrating an example of an image captured of the surface of a structure. [Figure 6] FIG. 4 is a diagram showing an example of first detection data. [Figure 7] 11A to 11C are diagrams for explaining an example of a combination information acquisition process. [Figure 8] 6A and 6B are diagrams showing examples of combination information and second detection data. [Figure 9] FIG. 13 is a diagram for explaining an example of analysis processing. [Figure 10]FIG. 11 is a diagram for explaining another example of the analysis process. [Figure 11] 11A and 11B are diagrams for explaining examples of analysis results for determining a position of interest. [Figure 12] FIG. 4 is a diagram illustrating a display screen displayed on an output device. [Figure 13] FIG. 13 is a diagram illustrating an example of a display image including a focus position. [Figure 14] FIG. 11 is a functional block diagram of an information processing apparatus according to a second embodiment. [Figure 15] 10 is a flowchart illustrating a control process according to a second embodiment. [Figure 16] FIG. 13 is a diagram showing an example of accepting a selection of combination information. [Figure 17] FIG. 13 is a functional block diagram of an information processing apparatus according to a third embodiment. [Figure 18] 13 is a flowchart illustrating a control process according to a third embodiment. [Figure 19] FIG. 4 is a diagram for explaining a learning data collection process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0009] <First embodiment> Hereinafter, an embodiment will be described in which the information processing device of the present invention is applied to a computer device used for inspecting infrastructure structures such as concrete structures. In the first embodiment, an example will be described in which the computer device operates as an information processing device and is configured to display the vicinity of an image position where the detection target is likely to exist when displaying the image and a detection result of detecting a detection target from an image of the inspection target.

[0010] In this embodiment, the "inspection target" is a structure made of concrete, etc., to be inspected, such as a highway, a bridge, a tunnel, or a dam. The information processing device performs a detection process to detect the presence or absence and state of a detection target, such as a crack, using an image captured by a user of the structure to be inspected. In addition, in the case of a concrete structure, the "detection target" is, for example, cracks, chalk, lifting, and spalling of the concrete. Other examples of the "detection target" include efflorescence (whitening phenomenon), exposed rebar, rust, water leakage, dripping, corrosion, damage (missing parts), cold joints, deposits, junk, etc.

[0011] <Hardware configuration> The hardware configuration of an information processing device of the first embodiment will be described with reference to FIG. 1(a). FIG. 1(a) is a block diagram showing the hardware configuration of an information processing device 100 of the first embodiment. In the first embodiment, a computer device operates as the information processing device 100. Note that the processing of the information processing device of this embodiment may be realized by a single computer device, or may be realized by distributing each function among multiple computer devices as necessary. The multiple computer devices are connected to each other so as to be able to communicate with each other.

[0012] The information processing device 100 includes a control unit 101, a non-volatile 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.

[0013] The control unit 101 includes an arithmetic processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) that controls the entire information processing device 100. The control unit 101 may have an arithmetic processor such as a GPU (Graphics Processing Unit) or a QPU (Quantum Processing Unit) in addition to or instead of a CPU.

[0014] The non-volatile memory 102 is a ROM (Read Only Memory) that stores programs executed by the processor of the control unit 101 and parameters required for executing the programs. Here, the programs refer to programs for executing the processes of each embodiment described below. Specifically, the non-volatile memory 102 stores an OS (operating system), which is basic software executed by the control unit 101, and applications that work with the OS to realize applied functions.

[0015] In this embodiment, the non-volatile memory 102 stores an application that enables the information processing device 100 to realize control processing. The control processing of the information processing device 100 in this embodiment is realized by reading software provided by the application. It is assumed that the application has software for utilizing basic functions of the OS installed in the information processing device 100. The OS of the information processing device 100 may have software for realizing the control processing in this embodiment.

[0016] The work memory 103 is a RAM (Random Access Memory) that temporarily stores programs and data supplied from an external device, etc. The work memory 103 holds data obtained by executing the control process shown in FIG.

[0017] The storage device 104 includes a memory card including a semiconductor memory and a magnetic disk, an SSD (Solid State Drive), a hard disk, etc. The storage device 104 also includes a storage medium including a disk drive that reads / writes data from / to an optical disk such as a DVD or a Blue-ray Disc. The storage device 104 may be an internal device built into the information processing device 100, or may be an external device detachably connected to the information processing device 100.

[0018] The input device 105 is an operating member such as a mouse, a keyboard, a touch panel, etc., that accepts user operations. The input device 105 outputs to the control unit 101 an operation instruction accepted from the user.

[0019] The output device 106 is an example of a display unit, and is a display device such as a display or monitor composed of an LCD (Liquid Crystal Display) and an organic EL (Electro Luminescence) display, etc., and displays data held by the information processing device 100 and data supplied from an external device.

[0020] The network interface 107 is communicably connected to a network such as the Internet or a LAN (Local Area Network).

[0021] The system bus 108 connects the control unit 101, non-volatile memory 102, work memory 103, storage device 104, input device 105, output device 106, and network interface 107 of the information processing device 100 so as to enable data transmission and reception. The system bus 108 includes an address bus, a data bus, a control bus, and the like.

[0022] <Function block> Next, referring to FIG. 1(b), functional blocks of the information processing device 100 of the first embodiment will be described. FIG. 1(b) is a functional block diagram of the information processing device 100 of the first embodiment. The information processing device 100 has a storage unit 121, a management unit 122, an image acquisition unit 123, a first detection data acquisition unit 124, a combination information acquisition unit 125, a second detection data acquisition unit 126, an analysis unit 127, a focus position determination unit 128, a display control unit 129, and a database 130. Each function of the information processing device 100 is configured by hardware and / or software. For example, the control unit 101 may read out a program and parameters stored in the non-volatile memory 102 and expand them in the work memory 103, so that the information processing device 100 realizes each function or a part of the function. Note that each functional unit may be configured as a system consisting of one or more computer devices or server devices connected by a network. Moreover, when each of the functional units shown in FIG. 1(b) is configured by hardware instead of being realized by software, it is sufficient to have a circuit configuration corresponding to each of the functional units in FIG. 1(b).

[0023] The management unit 122 manages the registration, deletion, acquisition, and updating of data to be processed, such as image data, which is data on images of structures and the like, and detection data, which is stored in the storage unit 121.

[0024] The image acquisition unit 123 acquires image data to be processed from the storage unit 121 via the management unit 122 .

[0025] The first detection data acquisition unit 124 acquires first detection data related to a first detection target. The first detection target is a crack in concrete, chalk, lifting, spalling, exposed rebar, efflorescence, water leakage, and the like.

[0026] The combination information acquisition unit 125 performs a process of acquiring combination information of detection targets stored in the database 130 .

[0027] The second detection data acquisition unit 126 acquires second detection data related to a second detection target that is highly related to the first detection target based on the combination information from the storage unit 121 via the management unit 122. The second detection target is any one of the first detection targets, such as chalk, exposed rebar, spalling, and cracks.

[0028] The analysis unit 127 analyzes the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information. Specifically, the analysis unit 127 sets an analysis range for the first detection data based on the second detection data related to the second detection target, and analyzes the first detection target combined with the second detection target within the analysis range of the first detection data to generate an analysis result.

[0029] The attention position determining unit 128 determines an attention position on the image based on the analysis result.

[0030] The display control unit 129 creates display data, which is data for a display screen, using the selected image, the first detection data, and the focus position, and outputs the display data to the output device 106, causing the output device 106 to display the display screen.

[0031] The processing and functions of the image acquiring unit 123, the first detected data acquiring unit 124, the combination information acquiring unit 125, the second detected data acquiring unit 126, the analyzing unit 127, the focus position determining unit 128, and the display control unit 129 will be described in detail later.

[0032] <Description of combination information> With reference to FIG. 2, the combination information used in the control process of the information processing device 100 of this embodiment will be described. FIG. 2 is an example of a table showing the combination information. The combination information is information showing a combination of detection targets that are likely to be present in the same position (i.e., have a high relationship) among detection targets such as cracks present on the wall surface of the inspection target. In this embodiment, the combination information is used to find an image position where the detection target is likely to be present, and is used for checking the detection results. The combination information is stored in the database 130 in advance. The combination information acquisition unit 125 acquires and uses the combination information from the database 130. The combination information stored in the database 130 may be any combination information. The database 130 may be stored in an external storage device such as a server connected via a network.

[0033] For example, in a close-up visual inspection at an on-site site, an inspector may mark the vicinity of a crack that occurs on the wall surface of the inspection target with chalk. Based on this knowledge, a combination of cracks and chalk can be stored as one of the combination information to be stored in the database 130. Furthermore, when the wall surface of the inspection target structure deteriorates, the surface may peel off and cause spalling. As this deterioration progresses, rebar inside the inspection target may become exposed around the spalling. Based on this knowledge, a combination of spalling and exposed rebar can be stored as one of the combination information to be stored in the database 130.

[0034] FIG. 2(a) shows combination information 201 of detection targets present on the wall surface of a bridge as an example of combination information. The combination ID of the combination information 201 is identification information for identifying multiple pieces of combination information. In the combination information 201, the confirmation target column and the reference target column indicate a combination of detection targets that are highly related to each other. The detection target of the reference target column is an object that is likely to exist in the vicinity of the detection target of the confirmation target column. For example, the detection target of the confirmation target column is the first detection target, and the detection target of the reference target column is the second detection target. In this way, one method of realizing this is to store the combination information in table format in the database 130.

[0035] In the combination information in this embodiment, there are cases where it is desired to limit the detection targets to a portion of the detection targets. For example, all cracks and chalks of a certain length or more may be set as the combination information. In such cases, a remarks column can be provided as in combination information 211 of FIG. 2(b) to enter the conditions for narrowing down the detection targets. Also, as in combination information 221 of FIG. 2(c), analysis conditions in the analysis unit 127 may be associated with the combination information. In this way, any information related to the analysis process can be associated with the combination information. The analysis process of the analysis unit 127 will be described later.

[0036] <Explanation of images and detection data> Next, images and detection data used in the control process of the information processing device 100 of this embodiment will be described with reference to Fig. 3. In image inspection, it is preferable to manage images of the wall surface of a structure in association with drawings.

[0037] FIG. 3(a) shows a state in which an image 311 of a bridge wall, as an example of an infrastructure structure, has been pasted onto a drawing 300. The drawing 300 has drawing coordinate axes 301 with point 302 as the origin. The position of the image on the drawing is defined by the coordinates of the vertex at the top left of the image. For example, the coordinates of the image 311 are the position of vertex 312 (X 312 , Y 312 ), and the image is stored in the storage unit 121 together with the coordinate information. In this embodiment, the images used in the image inspection of infrastructure structures are large in size because they are taken at a high resolution (e.g., 1 mm per pixel) so that minute cracks and the like can be confirmed. For example, image 311 in FIG. 3(a) is an image taken of a bridge deck measuring 20 m x 10 m. When the image resolution is 1.0 mm per pixel (1.0 mm / pixel), the image size of image 311 is 20,000 pixels x 10,000 pixels. Image 311 taken at a high resolution contains many detection targets of various sizes, such as cracks, chalk, exposed rebar, corrosion, water leakage, and efflorescence.

[0038] However, since it is difficult to show all detection targets on paper, only a portion of the detection targets are shown on paper. Detection data is information that records the results of AI using machine learning or deep learning to detect cracks on concrete wall surfaces from images, or the results of human input. Detection data may include detection errors by AI and input errors by humans, so there may be detection omissions.

[0039] The present embodiment aims to find such missed detections. In explaining the present embodiment, it is assumed that detection data is managed in association with a drawing. Fig. 3(b) shows a state in which detection data 321 corresponding to an image 311 is pasted on a drawing 300 at the same position as the image 311. The detection data 321 includes a large amount of detection data, including detection targets that are not displayed on the paper.

[0040] The position of each detection data in the detection data 321 on the drawing is defined by the pixel coordinates constituting the detection data. FIG. 3(c) shows an example of a detection data table 331 showing the data structure of the detection data. In the detection data table 331, the coordinate column is an attribute value showing a plurality of coordinates constituting the detection data. For example, it shows that the crack C001 is represented by continuous pixels consisting of n points from (Xc001_1, Yc001_1) to (Xc001_n, Yc001_n). In this way, in this embodiment, the detection data is represented by pixels. The detection data may be represented by vector data such as polylines and curves consisting of multiple points. When the detection data is represented by vector data, the data capacity is reduced and the representation is more simplified. An example of detection data other than cracks is rebar exposure with ID T001. When the rebar exposure is represented by a polyline, it becomes a detection having an area surrounded by the polyline. The attribute information held by the detection data is not limited to the attribute information shown in the detection data table 331, and other attribute information may be held.

[0041] In the control process of this embodiment, the information processing apparatus 100 can also use an image and detection data that are managed in association with each other. By associating the image and the detection data, the information processing apparatus 100 can determine the positional relationship between the detection data and the image without referring to the drawings.

[0042] <Control Process> Referring to FIG. 4, the control process of the information processing apparatus 100 in this embodiment will be described. The process in FIG. 4 is realized by the control unit 101 of the information processing apparatus 100 shown in FIG. 1(a) expanding and executing the program stored in the non-volatile memory 102 in the work memory 103, controlling each component shown in FIG. 1(a), and operating as each functional unit shown in FIG. 1(b). Hereinafter, the process of this embodiment will be described according to the flowchart of FIG. 4. In addition, each step (step) will be described with an S added at the beginning of those reference numerals.

[0043] <S401: Image Acquisition Process> The image acquisition unit 123 performs a process of acquiring an image to be processed. In this embodiment, the image stored in the storage unit 121 is acquired via the management unit 122. FIG. 5 shows an image 501 of a concrete wall surface of a bridge taken as an example of an inspection target. On the image 501, there are cracks 502, 503, 504, 505, and chalks 506, 507, 508, and 509. Note that the image 501 is an image taken at a high resolution and is very large in size. Therefore, there are also many other detection targets on the image 501, but it is difficult to represent all the detection targets on the drawing. Therefore, the detection targets shown in the drawing are limited to a part.

[0044] <S402: First Detection Data Acquisition Process> The first detection data acquisition unit 124 acquires first detection data of a first detection target corresponding to the image acquired in S401. FIG. 6 shows an example of the first detection data 601 when cracks are the first detection target. In the first detection data acquisition process, the first detection data acquisition unit 124 may use a method of displaying a screen for receiving a user's operation instruction and allowing the user to select the first detection data to be acquired. For example, as a method of allowing the user to select, the first detection data acquisition unit 124 displays a list of detection targets on the output device 106 and allows the detection target to be selected therefrom. Then, the first detection data acquisition unit 124 sets the selected detection target as the first detection target, and subsequently acquires the first detection data. As another method of allowing the user to select, the first detection data acquisition unit 124 may use a method of previously acquiring detection data of a plurality of detection targets, displaying them on the output device 106, and allowing some of the detection data to be selected as the first detection data. In this way, the first detection data acquisition unit 124 may acquire the first detection data according to the user's operation instruction.

[0045] <S403:Combination information acquisition process> The combination information acquisition unit 125 acquires combination information indicating a combination of a second detection target highly related to the first detection target from the database 130. FIG. 7(a) shows an example of the combination information 701 with a high degree of relevance among the detection targets occurring on the concrete wall surface of the bridge. In this combination information 701, the combination information acquisition unit 125 acquires the row in which the value of the confirmation target column is the same as the first detection target. As an example, FIG. 7(b) shows the combination information 711 with the combination ID "Id71" acquired when the first detection target is a crack. In this way, the combination information acquisition unit 125 can acquire the combination information of the second detection target highly related to the first detection target.

[0046] In the process of S403, a plurality of pieces of combination information may be acquired. For example, FIG. 7(c) shows the combination information 721 acquired from the combination information 701 with the exposed reinforcement as the first detection target. In this case, the combination information acquisition unit 125 can handle it by executing the processes after S404 for the number of rows of the combination information 721.

[0047] <S404: Second Detection Data Acquisition Process> The second detection data acquisition unit 126 performs a process of acquiring second detection data of a second detection target set based on the combination information. In the present embodiment, the second detection data acquisition unit 126 sets the second detection target based on the combination information acquired in S403, and acquires the second detection data of the second detection target. This second detection data acquisition process will be described with reference to FIG. 8.

[0048] FIG. 8(a) shows an example of the combination information 801 acquired in S403. The second detection data acquisition unit 126 refers to the reference target column of this combination information 801, and sets "chalk" as the second detection target. Then, the second detection data acquisition unit 126 acquires, as the second detection data, the data obtained by detecting the chalk, which is the second detection target on the image acquired by the image acquisition unit 123. FIG. 8(b) shows an example of the second detection data 811. In this way, the second detection data acquisition unit 126 can acquire the second detection data based on the combination information.

[0049] FIG. 8(c) shows an example of the combination information 821 including the narrowing-down condition of the second detection data. As shown in FIG. 8(c), the combination information 821 acquired in the process of S403 may include the narrowing-down condition of the second detection data. In this case, the second detection data acquisition unit 126 may acquire, as the second detection data, the chalk data with an actual dimension length of 0.01 m or more based on the remarks column of the combination information 821. In this way, the second detection data acquisition unit 126 can narrow down the detection data to be acquired as the second detection data.

[0050] <S405: Analysis Process> In S405, the analysis unit 127 sets an analysis range and performs a process of analyzing the first detection data within the analysis range. The analysis process in this embodiment will be described with reference to FIG. 9. FIG. 9(a) shows the first detection data to be analyzed. FIG. 9(b) shows the second detection data for setting the analysis range. The first detection data 901 in FIG. 9(a) and the second detection data 911 in FIG. 9(b) are examples of data in which cracks and chalk are detected in the same image, respectively, and are associated with the same position on the drawing.

[0051] When performing the analysis process, the analysis unit 127 first sets an analysis range. As a method for setting the analysis range, the analysis unit 127 may use the second detection data. Specifically, the analysis unit 127 selects one piece of chalk data 912 from the second detection data 911. Then, the analysis unit 127 sets a dotted line range 913 surrounding the chalk data 912. The size of the range 913 may be changed to any size as long as the range includes the coordinate information of the chalk data 912. For example, the analysis unit 127 may set the range surrounding the chalk data 912 as an initial range, and expand it by any size in the X-axis and Y-axis directions on the drawing. Next, the analysis unit 127 superimposes the range 913 on the first detection data 901. The analysis unit 127 sets the range 903 obtained by this superposition as the analysis range. In this way, the analysis unit 127 can set the analysis range using the second detection data.

[0052] Next, the analysis unit 127 performs an analysis process on the first detection data included in the analysis range. Specifically, the analysis unit 127 selects crack data 902 included in the range 903 from the first detection data 901. Then, the analysis unit 127 calculates the number of selected crack data and the total length of the selected crack data as an analysis process for the crack data 902. FIG. 9(c) shows an example of the analysis result 921. The detection data column of the analysis result 921 is a column that identifies the crack data 902 in the analysis range. The detection extension column and the detection data number column indicate the total length and the number of crack data in the analysis range, respectively. In addition, the analysis result 921 has a reference data column and an analysis range coordinate column as information related to the analysis range. The reference data column indicates a column of identification information that identifies the chalk data 912 selected as the reference for the analysis range. The analysis range coordinate column indicates coordinate information on the drawing of the range 903. Thereafter, the analysis unit 127 selects one piece of chalk data different from the chalk data 912 from the second detection data 911, and sets the analysis range and performs the analysis process in a similar manner. The analysis unit 127 repeatedly performs this process on all the chalk data in the second detection data 911, thereby obtaining a plurality of analysis results corresponding to each piece of chalk data.

[0053] The analysis unit 127 may use another method as a method for setting the analysis range using the second detection data. FIG. 10(a) shows an example of the second detection data 1001 when a crack is set as the second detection target. The analysis unit 127 may detect a closing crack based on the second detection data 1001 and set the analysis range by superimposing a range 1002 including the closing crack on the first detection data. The analysis unit 127 may use, for example, the technology of Patent Document 2 as a method for detecting a closing crack. In this way, the analysis unit 127 may also set the analysis range by analyzing the second detection data 1001.

[0054] The analysis unit 127 may use yet another method as a method for setting the analysis range. For example, the analysis unit 127 may use the entire image to be processed as one analysis range. The analysis unit 127 may also divide the image to be processed into rectangular ranges of a certain size and use each divided range as an analysis range. When the analysis unit 127 sets the analysis range without using the second detection data, the same processing as the analysis processing of the first detection data in the analysis range may also be performed on the second detection data. Thereby, the attention position determination unit 128 can determine the attention position by also using the analysis result of the second detection data in the process of S406 described later.

[0055] As another analysis method, the analysis unit 127 can also use different analysis methods for each combination information. FIG. 10(b) shows an example of combination information 1011 associated with analysis conditions. The analysis unit 127 sets the analysis range based on the reference data column and the analysis range column of the combination information 1011, and outputs the description content of the output item column as the analysis result. As described with reference to FIG. 2(c), the combination information associated with the analysis conditions may be stored in advance in the database 130. Thereby, in the process of S403, the combination information acquisition unit 125 can acquire the combination information associated with the analysis conditions. Therefore, the analysis unit 127 can use different analysis methods for each combination information.

[0056] <S406: Attention Position Determination Process> The attention position determination unit 128 performs a process of determining an attention position on an image based on the analysis result by the analysis unit 127. The attention position is, for example, a position within a plurality of analysis ranges where the second detection target is detected but the first detection target associated with the second detection target is not detected. In other words, the attention position is a position within the analysis range where there is a high possibility of a detection omission that the user or the like should pay attention to. The attention position determination process will be described with reference to FIG. 11. FIG. 11(a) is an example of the analysis result 1101 obtained in S405. First, the attention position determination unit 128 performs a process of narrowing down the analysis results used in the attention position determination process from the analysis result 1101. As an example of a narrowing down method, the attention position determination unit 128 may determine whether the number of detection data in the analysis result is less than a reference value or not, and narrow down the results based on the determination result. Equation 1 shows an example of a determination formula.

[0057]

number

[0058] In formula 1, parameter Dc is an arbitrary constant, and parameter D is the number of detection data in the analysis result. Here, in explaining this embodiment, parameter Dc is set to "1". Note that parameter Dc may be the number of second detection data within the target analysis range, that is, the number of second detection targets detected within the analysis range. This makes it possible to deal with a case where multiple chalks are detected within the analysis range and one of the corresponding multiple cracks is not detected. The attention position determination unit 128 selects one analysis result from the analysis results 1101, substitutes the number of detection data of the selected analysis result into formula 1, and determines whether formula 1 is satisfied. FIG. 11(b) shows an analysis result 1111 that satisfies formula 1 among the analysis results 1101.

[0059] Next, the attention position determination unit 128 determines the attention position based on the analyzed result 1111 after narrowing down. As a method for determining the attention position, among the analyzed results 1111, there is a method of calculating the barycentric coordinates of the reference data D011 that is the reference of the analysis range and setting the barycentric coordinates as the attention position. The reference data is a part of the second detection data and has coordinate information on the drawing. Therefore, the attention position determination unit 128 can calculate the barycentric coordinates on the drawing based on the reference data. As another method for determining the attention position, the attention position determination unit 128 may calculate the center coordinates of the analysis range and set the center coordinates as the attention position. In this way, the attention position determination unit 128 can determine the attention position on the image based on the analyzed result.

[0060] Although an example of using the number of detection data as a method for narrowing down the analyzed result has been described, the attention position determination unit 128 may use another method. For example, the attention position determination unit 128 may perform narrowing down based on the determination result of whether the detected area in the analyzed result is less than a reference value. In this case, the attention position determination unit 128 may directly use Equation 1 and set the reference value of the detected area for the parameter Dc, and substitute the value of the detected area of the analyzed result for the parameter D. In this way, the attention position determination unit 128 can narrow down the analyzed result using any information related to the analyzed result.

[0061] The narrowing-down process of the analyzed result by the attention position determination unit 128 can also be omitted. In this embodiment, by omitting the narrowing-down process, the attention positions corresponding to all the analyzed results can be obtained. Therefore, in the display process of S407 described later, not only the image positions with a high possibility of detection omission but also the image positions with a high possibility of the presence of the detection target can be confirmed without omission.

[0062] <S407: Display Process> In S407, the display control unit 129 creates display data for display as an image based on the partial image near the attention position and outputs it to the output device 106. The output device 106 displays the display image based on the display data.

[0063] A display screen 1201 in Fig. 12(a) is an example of a screen displayed based on display data created by the display control unit 129. Here, the first detection target on the display screen 1201 is a crack, and the second detection target is chalk. The user checks the crack detection results on the display screen 1201. A list of focus positions 1202 on the display screen 1201 is a list of focus positions obtained in the processing of S406.

[0064] In this embodiment, the display control unit 129 displays on the output device 106 an enlarged view 1205 of the vicinity of a focus position selected by the user from the focus position list 1202. The enlarged view 1205 is a view in which crack detection data and chalk detection data are superimposed on a partial image in the vicinity of a selected focus position 1204. In this enlarged view 1205, chalk data 1208 corresponding to a chalk 1207 on the image is present, but crack data corresponding to a crack 1206 on the image is not present. In this manner, the focus position indicates an image position where there is a high possibility of a detection miss related to the first detection target. Therefore, by utilizing the information on the focus position, the user can efficiently check the detection results.

[0065] The dotted range 1209 is the analysis range related to the analysis result used in the attention position determination process. In this way, the display control unit 129 visualizes the analysis range, thereby making it easier for the user to understand the image range to be confirmed. In addition, the user can individually switch between display and non-display of each detection data and partial image included in the enlarged view 1205 by selecting a display target with the display switch 1203. Note that the user can move or deform the thick frame 1216 of the overall view 1215 according to various operations of the input device 105. This allows the display control unit 129 to freely switch the display range and display position of the display image based on the display data displayed as the enlarged view 1205.

[0066] In the display process of the display control unit 129, the combination information acquired in the process of S403 can also be displayed on the screen. The display screen 1211 in FIG. 12(b) is an example when the first detection target is spalling and there are multiple second detection targets (cracks and exposed rebars). The user checks the detection result of spalling on the display screen 1211. The combination information list 1212 on the display screen 1211 is the combination information acquired in S403. By displaying the combination information on the display screen 1211, the user can easily recognize the combination information used to find the attention position. In addition, the display control unit 129 may switch the attention position list 1214 based on the combination information. Specifically, when the user presses the update button 1213, the display control unit 129 switches the display of the attention position list 1214 according to the selection of the combination information list 1212. In this way, the display control unit 129 may link the selection information of the combination information list 1212 with the display of the attention position list 1214.

[0067] The attention position list 1214 on the display screen 1211 shows a state in which the ascending sorting of the peeling detection area has been applied. In this way, the display control unit 129 may change the order of the attention position list 1214. Furthermore, when there are many attention positions, it takes time for the user to check all of the attention positions. Therefore, the display control unit 129 provides a filtering function for filtering and displaying only the top five positions with the least number of peeling detection data from the attention position list, thereby enabling the user to efficiently check the attention positions.

[0068] In the process of S407, the display control unit 129 may display the entire image including the attention position. After performing the attention position determination process for many images using this embodiment, in the process of S407, the display control unit 129 makes it possible to display the entire image including the attention position. In this way, the display control unit 129 makes it possible to pick up images that are likely to include a detection omission of the first detection target, thereby allowing the user to efficiently narrow down the images to be confirmed.

[0069] When there are many detection results near the focus position, it becomes difficult for the user to view the image near the focus position. In such a case, in the process of S407, the display control unit 129 may display only the detection data related to the analysis result used in the focus position determination process, and may hide the other detection data. Figures 13(a) and 13(b) respectively show examples of a display image 1301 and a display image 1311 when a peeling is set as the first detection target, and a crack with a maximum width of 1 mm or more is set as the second detection target.

[0070] The display image 1301 shows a state in which the peeling detection data and the crack detection data are superimposed on the image to be processed. The image to be processed of the display image 1301 includes a plurality of cracks in addition to the peeling 1302 used in the attention position determination process, the crack 1303 with a maximum width of 1 mm or more, and the crack data 1304. The display image 1311 is different from the display image 1301 in that only the crack data 1304 related to the attention position determination process is displayed among the plurality of crack data. The display image 1311 makes it easier for the user to visually recognize the peeling 1302, the crack 1303, and the crack data 1304 on the image compared to the display image 1301. In this way, the display control unit 129 may display only the detection data related to the attention position determination process and hide the other detection data.

[0071] <Modification of the first embodiment (utilization of correction results)> In the first embodiment, the analysis unit 127 can also use detection data corrected by the user. Specifically, the first detection data acquisition unit 124 acquires the correction history of the first detection data together with the first detection data of the first detection target from the storage unit 121. The first detection data acquisition unit 124 corrects the first detection data based on the correction history. In the subsequent analysis process, the analysis unit 127 performs the analysis process using the corrected first detection data. Other processes are the same as those in the first embodiment. The attention position thus obtained indicates a position where there is a high possibility of detection failure even after the user corrects the first detection data, so it is desirable to check it. Therefore, the first detection data acquisition unit 124 uses the correction history of the detection data, so that the analysis unit 127 can determine the image position where there is a high possibility of detection failure.

[0072] <Modification of the first embodiment (use of user operation history)> In the first embodiment, the information processing device 100 can also utilize the user's operation history including a browsing history indicating browsing of the detection result. Specifically, the first detection data acquisition unit 124 acquires the detection data of the first detection target and the user's browsing history of the processing target image. In the subsequent attention position determination process, the attention position determination unit 128 may determine the attention position on the image based on the operation history including the browsing history. For example, the attention position determination unit 128 determines the attention position from within an image range that the user did not view. In this way, the attention position determination unit 128 can determine, as the attention position, a position that is likely to be overlooked by the user during the detection result confirmation work by using the user's operation history. Note that the timing for using the user's browsing history does not have to be the attention position determination process. For example, in the second detection data acquisition process, the second detection data acquisition unit 126 may acquire the second detection data from within a range that the user did not view. Also, in the analysis process, the analysis unit 127 may exclude the range that the user viewed from the analysis range. In this manner, the information processing device 100 can use the user's browsing history in various ways.

[0073] According to the first embodiment described above, the information processing device 100 analyzes the first detection data indicating the result of the first detection target being detected from the image based on the combination information of the first detection target such as a crack and the second detection target such as chalk that is likely to be present around the first detection target and is related to the first detection target. This allows the information processing device 100 to suppress the omission of detection of the first detection target compared to the case where the first detection target is detected and analyzed alone.

[0074] In the information processing device 100, the display control unit 129 generates display data for a display image based on the analysis result of the first detection data, outputs the display data to the output device 106, and causes the display image to be displayed on the output device 106. This enables the information processing device 100 to provide the user with a display image based on the analysis result with few detection omissions.

[0075] In the information processing device 100, the second detection data acquisition unit 126 acquires second detection data that is a detection result of a second detection target related to the first detection target. This allows the information processing device 100 to analyze the first detection data based on the second detection data, so that even if the first detection target is not detected, useful information can be provided to the user by detecting the second detection target.

[0076] In the information processing device 100, the analysis unit 127 sets an analysis range in the image based on the second detection data, and analyzes the first detection data in the analysis range to generate an analysis result. This enables the information processing device 100 to reduce the analysis range and reduce the processing load, while efficiently suppressing the omission of detection of the first detection target.

[0077] In the information processing device 100, the attention position determining unit 128 determines the attention position within the analysis range. This enables the information processing device 100 to reduce the range on which the user should pay attention, thereby reducing the checking burden on the user.

[0078] In the information processing device 100, the display control unit 129 determines a focus position within the analysis range on the image according to the combination information, and creates display data for a display screen in which a partial image near the focus position and the first detection data of the detection target are superimposed. This allows the information processing device 100 to generate a display screen with a partial image including a focus position with a high probability of detection omission that the user should pay attention to. The information processing device 100 then displays the display screen created by the display control unit 129 on an output device, thereby improving the efficiency of the user's work of checking the detection data.

[0079] In the information processing device 100, the display control unit 129 displays a display screen including combination information associated with the coordinates of the attention position. This allows the information processing device 100 to provide the user with the necessary division in detail.

[0080] In the information processing device 100, the display control unit 129 includes an enlarged view including the analysis range including the detection data on the display screen. This allows the information processing device 100 to provide the user with noteworthy detection data among the detection data analyzed within the analysis range.

[0081] The information processing device 100 sets cracks and chalk as the first detection target, and chalk as the second detection target. The information processing device 100 analyzes the first detection data based on combination information that associates cracks with chalk, so even if a crack is not detected, if chalk is detected, it can determine a position of interest within the analysis range of the first detection data, thereby reducing missed detection of cracks.

[0082] <Second embodiment> In the first embodiment, an example was described in which the attention position is automatically determined using combination information of detection targets. When there is a large amount of combination information of detection targets, a large calculation cost is required. Therefore, the information processing device 100 of the second embodiment displays the combination information of the detection targets on a screen in addition to the image and detection data of the processing target, and accepts the selection of combination information by the user. Then, the information processing device 100 performs the attention position determination process based on the combination information selected by the user. In this way, the information processing device 100 of the second embodiment can reduce the calculation cost by having the user select the combination information to be used in the attention position determination process.

[0083] The hardware configuration of the information processing device 100 according to the second embodiment is similar to that of the first embodiment shown in FIG. 1(a), and therefore illustration and description thereof will be omitted. FIG. 14 is a diagram showing an example of a functional block diagram of the information processing device 100 according to the second embodiment. The functional configuration of FIG. 14 differs from the functional configuration of the first embodiment shown in FIG. 1(b) in that a receiving unit 1401 is added. The receiving unit 1401 is a functional unit of the control unit 101, and performs processing to receive a user selection for the combination information output to the output device 106.

[0084] Fig. 15 is a flowchart showing a main flow of information processing executed by the information processing device 100 according to the second embodiment. In the flowchart of Fig. 15, in the steps having the same numbers as those in the flowchart of Fig. 4 described in the first embodiment, the same processing as in the first embodiment is executed.

[0085] In the case of the second embodiment, after acquiring the combination information in S403, the information processing device 100 proceeds to S1501. In the process of S1501, the display control unit 129 creates display data including the combination information acquired in S403, and causes the output device 106 to display a display image based on the display data. Then, in S1502, the receiving unit 1401 receives a selection of combination information from the user.

[0086] After that, in S404, the second detection data acquisition unit 126 acquires second detection data based on the selected combination information. After that, through the analysis process in S405 and the attention position determination process in S406, in S407, the display control unit 129 creates display data and outputs it to the output device 106. In S1503, when the information processing apparatus 100 determines that the browsing has ended, the process ends; otherwise, the process of S1502 is executed again. Hereinafter, the outlines of the processes of S1501 and S1502 will be described.

[0087] <S1501: Output Process of Display Data Including Combination Information> The display control unit 129 creates display data including the combination information acquired in S403 and outputs it to the output device 106. The screen 1601 in FIG. 16 is an example of a screen displaying a display image by the display data when cracks are the first detection target. The enlarged view 1602 on the screen 1601 is a part of the display image obtained by superimposing the image to be processed and the crack detection data, and there are cracks 1603 and crack data 1604 on the image. Also, the combination information list 1605 on the screen 1601 is an example of the combination information related to the first detection target acquired in S403. The user confidence level is an operation of selecting some combination information from the combination information list 1605 while confirming the image to be processed and the crack detection data on the screen 1601.

[0088] <S1502: Reception Process of Combination Information> The reception unit 1401 performs a process of receiving the selection of combination information by the user. For example, on the screen 1601 in FIG. 16, when the user presses the analysis execution button 1606, in response to the button press operation, the reception unit 1401 acquires the combination information selected by the user from the combination information list 1605. In the subsequent process of S404, using the combination information acquired by the reception unit 1401, the second detection data acquisition unit 126 acquires second detection data in the same manner as the process of the first embodiment.

[0089] If the user selects multiple pieces of combination information, the reception unit 1401 receives all of the selected pieces of combination information. In this case, similarly to the first embodiment, the processes of S404, S405, and S406 may be executed for each piece of combination information.

[0090] By using the method described above, the information processing device 100 performs attention position determination processing using the combination information selected by the user and accepted by the accepting unit 1401. This allows the information processing device 100 to generate a display image based on an analysis result based on the second detection target in accordance with the user's desire, while suppressing unnecessary calculation costs.

[0091] <Third embodiment> In the first embodiment, an example was described in which a position where there is a high possibility of a detection target such as a crack being missed is obtained as the attention position. This missed detection means that the attention position is located near a detection target that is difficult to detect with an existing AI model. Therefore, images near this attention position are effective data for updating the AI ​​model. Therefore, in the third embodiment, a method of collecting images near the attention position as learning data will be described.

[0092] The hardware configuration of the information processing device 100 according to the third embodiment is similar to that of the first embodiment shown in Fig. 1(a), and therefore illustration and description thereof will be omitted. Fig. 17 is a diagram showing an example of a functional block diagram of the information processing device 100 according to the first embodiment. The functional configuration of Fig. 17 differs from the functional configuration of the first embodiment shown in Fig. 1(b) in that a collection unit 1701 is added. The collection unit 1701 is a functional unit of the control unit 101, and performs processing to collect learning data.

[0093] FIG. 18 is a flowchart showing the main flow of information processing executed by the information processing apparatus 100 according to the third embodiment. In the flowchart of FIG. 18, in the steps assigned the same numbers as those in the flowchart of FIG. 4 described in the first embodiment, the same processing as that in the first embodiment is executed. In the case of the third embodiment, after the display control unit 129 outputs display data to the output device 106 in S407, the information processing apparatus 100 proceeds to the processing of S1801. In S1801, the collection unit 1701 performs a process of collecting learning data. Hereinafter, the outline of the processing of S1801 will be described.

[0094] <S1801: Learning Data Collection Process> In S1801, the collection unit 1701 sets a collection range in the image and performs a process of collecting at least one of the image within the collection range and the first detection data as learning data. In explaining the learning data collection process of this embodiment, cracks are the first detection target and chalk is the second detection target. FIG. 19(a) shows an example of an image 1901 near the attention position created in the process of S407. On the image 1901, there are cracks 1902 and chalk 1903 that are detection targets. An example of crack detection data 1911 corresponding to this image 1901 is shown in FIG. 19(b). There is no crack data corresponding to the crack 1902 on the crack detection data 1911. Thus, there is a high possibility that there are detection targets that could not be detected in the partial image near the attention position. The collection unit 1701 may set a collection range near the attention position like this image 1901 and collect the partial image, which is the image within the collection range, as learning data.

[0095] The method of setting the image range to be collected as learning data may be, for example, a method of setting the analysis range related to the analysis result used in the attention position determination process of S406. The collection unit 1701 can efficiently collect images that are likely to have missed detections by collecting partial images of the collection range corresponding to the analysis range. The collection unit 1701 may freely change the size of the collection range set based on the analysis range. For example, the collection unit 1701 can change the collection range by setting the analysis range as the initial collection range and expanding the collection range in any direction of the X-axis and Y-axis on the drawing. The collection unit 1701 can also reduce the collection range so as to include at least the analysis range. As a method of setting the collection range, the collection unit 1701 can also set the analysis range including the attention position as the collection range.

[0096] Furthermore, the collecting unit 1701 may determine the range of images to be collected as learning data based on a user's operation instruction as a method of setting the range. For example, in the process of S407, the display control unit 129 displays display data including a partial image near the focus position. The user inputs the range to be collected as learning data while changing the display position and the display range on the display data by operating an input device. The collecting unit 1701 extracts the partial image corresponding to the input range and collects it as learning data. In this way, the collecting unit 1701 may determine the range of images to be collected as learning data based on a user's operation instruction.

[0097] The collecting unit 1701 may collect the entire image including the attention position as the learning data as the image to be collected as the learning data. The collecting unit 1701 can collect images including the attention position on an image-by-image basis, thereby collecting images that are likely to have missed detections. The collecting unit 1701 may also collect detection data in the learning data collection process. The collected detection data can be used as the initial value of the teacher data. In this way, the collecting unit 1701 may collect the detection data together with the image.

[0098] The analysis unit 127 may set an analysis range based on a revision history of the first detection data by the user, and analyze the first detection data in the revised analysis range. For example, the analysis unit 127 may set an analysis range revised by the user from among the multiple set analysis ranges as a new analysis range, and analyze the first detection data in the analysis range again. In this case, the collection unit 1701 may collect, as learning data, the collection range set based on the revision history of the first detection data by the user in the learning data collection process.

[0099] The user checks the detection data of the first detection target while visually checking the partial image around the position of interest on the screen created in the process of S407. If a detection omission is found during the checking work, the user adds or corrects the detection data. FIG. 19(c) shows an example of crack detection data 1921 after the user adds to the crack detection data 1911 while visually checking the image 1901. The crack detection data 1921 has crack data 1922 added thereto in comparison with the crack detection data 1911. This crack data 1922 corresponds to the crack 1902 on the image 1901. That is, the crack 1902 is a crack that could not be detected by the detection process, and it is desirable to collect an image around this crack 1902. Specifically, the collection unit 1701 sets a range 1923 surrounding the crack data 1922, and extracts and collects a partial image corresponding to the range 1923 from the image 1901. The collecting unit 1701 may freely change the size of the range 1923 as long as the range includes the coordinates of the crack data 1922. The collecting unit 1701 may also collect the first detection data such as the crack data 1922 together with the partial image as learning data. This allows the information processing device 100 to reduce the cost of creating teacher data of the detection target for the collected images used as learning data. As described above, the collecting unit 1701 can collect learning data based on the correction history of the first detection data.

[0100] In the information processing device 100, the collection unit 1701 sets a collection range near a position of interest and collects partial images that are images in the collection range. This allows the information processing device 100 to use the partial images of interest as learning data, thereby enabling efficient collection of learning data.

[0101] In the information processing device 100, the collection unit 1701 collects images in a collection range set based on an analysis range set based on a correction history as partial images. As a result, the information processing device 100 sets a corrected range, i.e., a range where there is a high possibility of detection failure, as a collection range and collects partial images in the collection range as learning data, so that learning data can be collected efficiently.

[0102] In the information processing device 100, the collection unit 1701 collects the corrected first detection data together with the partial image as learning data, thereby enabling the information processing device 100 to efficiently learn based on the corrected first detection data.

[0103] (Other Examples) The present invention can also be realized by a process in which a program for implementing one or more of the functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) that implements one or more of the functions.

[0104] The above-described embodiments may be combined as appropriate. For example, the second embodiment may be combined with the third embodiment.

[0105] The disclosure of this specification includes the following information processing device, information processing method, and program. (Item 1) an image acquisition means for acquiring an image of a structure in which a detection target is present; a first detection data acquisition means for acquiring first detection data which is data relating to a plurality of detection targets detected from the image; a combination information acquisition means for acquiring combination information regarding a combination of a first detection target included in the plurality of detection targets and a second detection target related to the first detection target; an analysis means for analyzing the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information, and generating an analysis result; An information processing device comprising: (Item 2) A display control means for generating display data, which is data for a display screen, based on the analysis result and displaying the display screen on a display unit. 2. The information processing device according to item 1, further comprising: (Item 3) a second detection data acquisition means for acquiring second detection data which is data relating to the second detection target detected from the image based on the combination information; 3. The information processing device according to item 1 or 2, further comprising: (Item 4) a receiving unit for receiving a selection of the combination information, The second detection data acquisition means acquires the second detection data based on the selected combination information. 4. The information processing device according to item 3. (Item 5) The analysis means includes: setting an analysis range in the image that includes at least a portion of the second detection target based on the second detection data; Analyzing the first detected data in the analysis area to generate an analysis result 5. The information processing device according to item 3 or 4. (Item 6) a focus position determining means for determining a focus position within the analysis range based on the analysis result; 6. The information processing device according to item 5, further comprising: (Item 7) a display control means for generating a partial image including the target position from the image, generating display data which is data of a display screen including the partial image and the first detection data, and displaying the display screen on a display unit; 7. The information processing device according to item 6, further comprising: (Item 8) The display control means includes, on the display screen, the combination information that associates information about the focus position. 8. The information processing device according to item 7, (Item 9) A display control means for generating display data, which is data for a display screen, based on the analysis range and displaying the display screen on a display unit. 9. The information processing device according to any one of items 6 to 8, further comprising: (Item 10) The analysis means analyzes the first detection data in the analysis range based on a correction history of the first detection data by a user. 10. The information processing device according to any one of items 6 to 9, (Item 11) The attention position determining means determines the attention position on the image based on the analysis result and an operation history including a browsing history of the first detection data by a user. 11. The information processing device according to any one of items 6 to 10, (Item 12) a collection means for collecting partial images of the image within a collection range set based on the analysis range set based on the correction history; 11. The information processing device according to item 10, further comprising: (Item 13) The collecting means collects a partial image of the image within a collection range including the position of interest. 13. The information processing device according to item 12, (Item 14) The collecting means collects the first detection data included in the partial image. 14. The information processing device according to item 12 or 13. (Item 15) The first detection target is a crack and chalk present in the structure, The second detection target is the chalk associated with the crack by the combination information. 15. The information processing device according to any one of items 1 to 14, (Item 16) an image acquisition step of acquiring an image of a structure in which a detection target is present; a first detection data acquisition step of acquiring first detection data which is data relating to a plurality of detection targets detected from the image; a combination information acquisition step of acquiring combination information regarding a combination of a first detection target included in the plurality of detection targets and a second detection target related to the first detection target; an analysis step of analyzing the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information to generate an analysis result; An information processing method comprising: (Item 17) A program for causing a computer to function as each of the means of the information processing device according to any one of items 1 to 15.

[0106] The invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0107] 100...information processing device, 101...control unit, 121...memory unit, 122...management unit, 123...image acquisition unit, 124...first detection data acquisition unit, 125...information acquisition unit, 126...second detection data acquisition unit, 127...analysis unit, 128...attention position determination unit, 129...display control unit, 130...database.

Claims

1. an image acquisition means for acquiring an image of a structure in which a detection target is present; a first detection data acquisition means for acquiring first detection data which is data relating to a plurality of detection targets detected from the image; a combination information acquisition means for acquiring combination information regarding a combination of a first detection target included in the plurality of detection targets and a second detection target related to the first detection target; an analysis means for analyzing the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information, and generating an analysis result; An information processing device comprising:

2. A display control means for generating display data, which is data for a display screen, based on the analysis result and displaying the display screen on a display unit. The information processing apparatus according to claim 1 , further comprising:

3. a second detection data acquisition means for acquiring second detection data which is data relating to the second detection target detected from the image based on the combination information; The information processing apparatus according to claim 1 , further comprising:

4. a receiving unit for receiving a selection of the combination information, The second detection data acquisition means acquires the second detection data based on the selected combination information.

4. The information processing apparatus according to claim 3.

5. The analysis means includes: setting an analysis range in the image including at least a portion of the second detection target based on the second detection data; Analyzing the first detected data in the analysis area to generate an analysis result 4. The information processing apparatus according to claim 3.

6. a focus position determining means for determining a focus position within the analysis range based on the analysis result; The information processing apparatus according to claim 5, further comprising:

7. a display control means for generating a partial image including the target position from the image, generating display data which is data of a display screen including the partial image and the first detection data, and displaying the display screen on a display unit; The information processing apparatus according to claim 6, further comprising:

8. The display control means includes, on the display screen, the combination information that associates information about the focus position.

8. The information processing apparatus according to claim 7,

9. A display control means for generating display data, which is data for a display screen, based on the analysis range and displaying the display screen on a display unit. The information processing apparatus according to claim 6, further comprising:

10. The analysis means analyzes the first detection data in the analysis range based on a correction history of the first detection data by a user.

7. The information processing apparatus according to claim 6,

11. The attention position determining means determines the attention position on the image based on the analysis result and an operation history including a browsing history of the first detection data by a user.

7. The information processing apparatus according to claim 6,

12. a collection means for collecting partial images of the image within a collection range set based on the analysis range set based on the correction history; The information processing apparatus according to claim 10, further comprising:

13. The collecting means collects a partial image of the image within a collection range including the position of interest.

13. The information processing apparatus according to claim 12.

14. The collecting means collects the first detection data included in the partial image.

13. The information processing apparatus according to claim 12.

15. The first detection target is a crack and chalk present in the structure, The second detection target is the chalk associated with the crack by the combination information.

2. The information processing apparatus according to claim 1,

16. an image acquisition step of acquiring an image of a structure in which a detection target is present; a first detection data acquisition step of acquiring first detection data which is data related to a plurality of detection targets detected from the image; a combination information acquisition step of acquiring combination information regarding a combination of a first detection target included in the plurality of detection targets and a second detection target related to the first detection target; an analysis step of analyzing the first detection target of the first detection data based on the second detection target combined with the first detection target indicated by the combination information to generate an analysis result; An information processing method comprising:

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

Images