Image processing device, image processing method, and program

Abstract

To provide a technology that efficiently acquires a deformation detection result.SOLUTION: An image processing apparatus comprises: storage means that stores a first detection result which includes first deformation information detected in a first image of a first component folder of a subject and a second detection result which includes second deformation information detected in a second image of a second component folder being different from the first component folder; reception means that receives at least any of a first output condition which outputs the first detection result and the second detection results for each type of deformation associated with each of the first image and the second image, and a second output condition which outputs at least one of the first detection result and the second detection result; and output means that outputs a file which includes at least one of the first component folder including at least a portion of the first detection result, and the second component folder including at least a portion of the second detection result, on the basis of the reception result of the reception means.SELECTED DRAWING: Figure 2

Description

[Technical field]

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

[0002] Concrete surfaces of bridges and buildings may develop defects (cracks, water leakage, etc.) due to various factors. Since defects in structures caused by such defects need to be discovered and repaired early, inspection workers regularly visually inspect the concrete surface and mark the deformed areas directly with chalk to inspect them. After the inspection, the inspection results are written on a drawing based on the chalk marks, and an inspection report is created and submitted to the national or local government office. Recently, to save labor, a system has been introduced that automatically detects defects by inputting inspection images of concrete surfaces taken with a camera into the system. However, with this method, it is necessary to make a final check to see whether the detection accuracy of multiple defects automatically detected when creating the inspection report is the same as when inspected visually. In addition, it is necessary to correct the multiple detected deformed areas as necessary, and then submit photos and drawings that clearly indicate the deformed areas. Therefore, the procedure for creating an inspection report is complicated for inspection workers, and the amount of work involved in creating an inspection report remains large, just as it was before. Therefore, in Patent Documents 1 and 2, in order to reduce the burden of creating inspection reports, the inspection system is equipped with a function that checks multiple detected abnormality areas on the system and outputs them superimposed on the inspection image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7095418 [Patent Document 2] Patent Publication No. 2021-148606 Summary of the Invention [Problem to be solved by the invention]

[0004] In the inspection systems of Patent Documents 1 and 2, all detected abnormality data is superimposed on one inspection image and output. However, one detection result data may contain a mixture of accurate detection result data that matches the actual abnormality and inaccurate detection result data that does not match the actual abnormality. On the other hand, only accurate detection result data must be included in the inspection report. Therefore, in order to create an accurate inspection report, the user must extract only accurate data from the detection result data downloaded from the inspection system, or correct and edit inaccurate detection result data. Thus, one of the issues is that the workload of the inspection operator is still heavy.

[0005] Furthermore, when downloading the results of structure abnormality detection from the inspection system, the inspection results linked to the structure folder cannot be acquired all at once, so the inspection worker must select and acquire the detection result data to be included in the inspection report one by one on the inspection system. The structure folder is composed of part folders with parent-child relationships, such as the entire bridge, one pier, and part of the wall, and the inspection images are saved in this structure on the inspection system. However, since the detection result data is acquired for each execution result when downloading, the downloaded detection result data no longer maintains the parent-child relationship of the structure. Therefore, there is an issue that when creating an inspection report, it is difficult for the user to select parts and detection result data (images) to be used in the inspection report from folders that store various parts and a large amount of detection result data acquired from the inspection system.

[0006] Therefore, an object of the present invention is to provide a technique for efficiently acquiring abnormality detection results. [Means for solving the problem]

[0007] In order to achieve the object of the present invention, an image processing device according to one embodiment of the present invention is characterized in that it comprises: a storage means for storing a first detection result including first deformation information detected in a first image of a first parts folder of a subject, and a second detection result including second deformation information detected in a second image of a second parts folder different from the first parts folder; a reception means for receiving at least one of a first output condition for outputting the first detection result and the second detection result for each type of deformation contained in the first deformation information and the second deformation information, respectively, and a second output condition for outputting at least one of the first detection result and the second detection result; and an output means for outputting a file including at least one of the first parts folder including at least a portion of the first detection result and the second parts folder including at least a portion of the second detection result, based on the reception result of the reception means. Effect of the Invention

[0008] According to the present invention, a technique for efficiently acquiring abnormality detection results can be provided. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing the hardware configuration of an image processing apparatus according to a first embodiment. [Diagram 2] FIG. 1 is a block diagram showing the functional configuration of an image processing apparatus according to a first embodiment. [Diagram 3] 4A to 4C are views for explaining a method for uploading an inspection image to an image analyzing device according to the first embodiment. [Figure 4] 5A to 5C are views for explaining a method for downloading a detection result image to a user terminal according to the first embodiment. [Diagram 5] 11 is a screen showing a list of abnormality detection results according to the first embodiment. [Figure 6] 11 is a detailed display screen of a deformation detection result according to the first embodiment. [Figure 7] 4 is a view for explaining a hierarchical structure of a storage folder for an inspection image according to the first embodiment. FIG. [Figure 8] 5A to 5C are views for explaining the relationship between an inspection image and an execution ID according to the first embodiment. [Figure 9] FIG. 2 is a schematic diagram of an inspection report according to the first embodiment. [Figure 10] FIG. 13 is a diagram showing a detection result image displaying anomaly detection results for each anomaly type according to the first embodiment. [Figure 11] FIG. 13 is a diagram showing a detection result image displaying all abnormality detection results according to the first embodiment. [Figure 12] FIG. 2 is a view for explaining a hierarchical structure of folders in a zip file according to the first embodiment. [Figure 13] FIG. 11 is a diagram for explaining a setting screen for acquiring a deformation detection result according to the first embodiment. [Figure 14] FIG. 11 is a diagram for explaining a setting screen for collectively acquiring abnormality detection results according to the first embodiment. [Figure 15] 6 is a flowchart illustrating a process in which the image processing device according to the first embodiment transmits a detection result image associated with one execution ID. [Figure 16] 11 is a flowchart for explaining a process in which the image processing device according to the first embodiment transmits all detection result images in a structure folder at once. [Figure 17] 13A and 13B are diagrams illustrating a screen for making settings for acquiring a detection result image according to the second embodiment. [Figure 18] 17 is a flowchart for explaining that an image processing apparatus according to a second embodiment executes a process relating to a combination of the process in FIG. 15 and the process in FIG. 16. [Figure 19] 13A and 13B are diagrams illustrating a screen for setting acquisition conditions for a detection result image for each part folder of a structure according to the third embodiment. [Figure 20] FIG. 13 is a diagram illustrating a setting screen for acquiring a detection result image having a deformation width and deformation area equal to or greater than a certain level in the fourth embodiment. [Figure 21] 13A and 13B are diagrams illustrating a filtering setting screen for a detection result image according to the fifth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] 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.

[0011] (First embodiment) In the first embodiment, when obtaining the abnormality detection result of a structure, the user sets the acquisition conditions for the inspection image and the abnormality detection result. According to the first embodiment, the user can easily find the detection result image in which the abnormality that needs to be reported in the inspection report is accurately detected. In addition, the user can quickly determine whether the detection result image is suitable for attachment to the inspection report by closely examining the width and length of the crack and the water leakage range for each abnormality detection result. In addition, by setting the execution results to be obtained for each inspection image collectively, the detection result image can be obtained while maintaining the folder structure in which the inspection images are saved on the image processing device. As a result, the user can quickly obtain the detection result image to be used in the inspection report, and the time required to create the inspection report can be reduced compared to the conventional method.

[0012] (Definition of terms) In this specification, "deformation" includes cracks and water leakage that occur on concrete surfaces due to damage, deterioration, and other factors of structures such as highways, bridges, tunnels, and dams. "Cracks" refer to linear damage with a starting point, an end point, a length, and a width that occurs on the walls of structures due to aging or the impact of earthquakes. "Water leakage" refers to cracks that occur in concrete due to the effects of rain, etc., and water seeping in through gaps in the concrete, i.e., water leaking.

[0013] <Hardware configuration> FIG. 1 is a diagram showing the hardware configuration of an image processing apparatus according to the first embodiment.

[0014] The image processing device 101 includes a control unit 111, a volatile memory 112, a non-volatile memory 113, a storage device 114, an input device 115, an output device 116, a communication device 117, and a system bus 118. Note that a user terminal 102, which will be described later, may have the same hardware configuration as the image processing device 101.

[0015] The control unit 111 includes a central processing unit (CPU) that controls the entire image processing device 101, and an arithmetic processor such as an MPU.

[0016] The volatile memory 112 is a RAM (Random Access Memory) that temporarily stores programs and data supplied from an external device (not shown) or the like.

[0017] The non-volatile memory 113 is a ROM (Read-Only Memory) that stores programs and parameters executed by the processor of the control unit 111 .

[0018] The storage device 114 includes internal devices such as a hard disk and a memory card built into the image processing device 101, or external devices such as a hard disk and a memory card that are detachable from the image processing device 101. The storage device 114 includes a hard disk and a memory card constituted by a semiconductor memory, a magnetic disk, etc. The storage device 114 also includes a storage medium constituted by a disk drive that reads / writes data from / to optical disks such as DVDs and Blue-ray Discs.

[0019] The input device 115 is an operation means such as a mouse, a keyboard, or a touch panel that accepts user operations, and outputs operation instructions to the control unit 111.

[0020] The output device 116 is a display device such as a display or monitor configured with an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display. The output device 116 displays data stored in the image processing device 101, data supplied from an external device (not shown), and acquisition condition setting data.

[0021] The communication device 117 is communicably connected to a network such as the Internet or a LAN (Local Area Network).

[0022] The system bus 118 includes an address bus, a data bus, and a control bus that enable each component of the image processing device 101 to send and receive data.

[0023] The non-volatile memory 113 records an OS, which is basic software executed by the control unit 111, and applications that cooperate with the OS to realize applied functions. In this embodiment, the non-volatile memory 113 also stores an operation program for the image processing device 101.

[0024] The processing of the image processing device 101 is realized by reading software provided by an application. The application has software for utilizing basic functions of the OS installed in the image processing device 101. The OS of the image processing device 101 may have software for realizing the processing in this embodiment.

[0025] <Functional configuration of image processing device> Next, an example of the functional configuration of the image processing device 101 will be described with reference to FIGS.

[0026] FIG. 2 is a diagram showing the functional configuration of the image processing device according to the first embodiment.

[0027] The image processing device 101 includes an image management unit 211 , an image storage unit 212 , an image analysis unit 213 , an image analysis result storage unit 214 , an image analysis result management unit 215 , and an image analysis result acquisition unit 216 .

[0028] The image management unit 211 displays a list of images (hereinafter also referred to as inspection images) of a structure and details of each inspection image on the output device 116, and deletes the inspection images.

[0029] The image storage unit 212 stores the inspection images.

[0030] The image analysis unit 213 performs image analysis related to anomaly detection on the inspection image using a learning model (AI) created by machine learning and deep learning. The image analysis unit 213 transmits the image analysis result (hereinafter also referred to as anomaly detection result) to the image analysis result storage unit 214.

[0031] The image analysis result storage unit 214 stores the image analysis results of the inspection image.

[0032] The image analysis result management unit 215 displays a list and details of the abnormality detection results stored in the image analysis result storage unit 214.

[0033] The image analysis result acquisition unit 216 edits the anomaly detection results stored in the image analysis result storage unit 214 based on the user's anomaly detection result acquisition settings, and transmits (outputs) a file containing the edited anomaly detection results to the user terminal 102.

[0034] The functional units in FIG. 2 are connected to each other via a system bus 221. The functions of the image processing device 101 are realized by hardware and software. The functional units may be configured as one or more computer devices or server devices, or may be configured as a system connected via a network. The user terminal 102 may also have the same functional configuration as the image processing device 101.

[0035] 3 is a diagram for explaining a method for uploading an inspection image to an image analysis device according to the first embodiment. The image analysis system includes an image processing device 101 and a user terminal 102. Note that the image analysis system may include a configuration in which the image processing device 101 and the user terminal 102 are integrated together.

[0036] A user uploads an inspection image 331 to the image processing device 101 via a user terminal 102. The inspection image 331 is an image of a structure photographed by the user. The user terminal 102 is a local PC of the user, but is not limited to this and includes, for example, a smartphone and a tablet terminal.

[0037] The image storage unit 212 stores the inspection image 331 received from the user terminal 102 in association with the structure folder 311 .

[0038] FIG. 4 is a diagram illustrating a method for downloading a detection result image to a user terminal according to the first embodiment.

[0039] The image analysis result storage unit 214 stores the detection result image 422 in association with the execution ID 421. The execution ID 421 is a management number that distinguishes the contents of the abnormality detection performed on the inspection image, and is illustrated as "1", for example.

[0040] The detection result image 422 includes abnormality information such as cracks 423 and water leakage 424, and an inspection image 425. The abnormality information such as cracks 423 and water leakage 424 includes dimensional information such as the position, size, and thickness in the inspection image 425. The dimensional information is, for example, information in which relative coordinates are expressed in millimeters based on the position of the detected object (i.e., the abnormality) in the inspection image 425 in a planar rectangular coordinate system. The image processing device 101 transmits (outputs) the detection result image 422 to the user terminal 102. As a result, the detection result image 422 is downloaded to the user terminal 102.

[0041] Fig. 5 is a list display screen of the abnormality detection result according to the first embodiment. Fig. 6 is a detailed display screen of the abnormality detection result according to the first embodiment.

[0042] The image analysis result management unit 215 has a screen 501 that displays a list of the anomaly detection results stored in the image analysis result storage unit 214, as shown in Figure 5, and a screen 601 that displays each anomaly detection result in detail, as shown in Figure 6.

[0043] 5, a screen 501 has an execution ID 511, a detection date and time 512, a status 513, a result acquisition button 514 for acquiring the abnormality detection results linked to the execution ID, a collective result acquisition button 515 for acquiring all abnormality detection results in the structure folder, and a result detail display button 516. The abnormality detection results may be arranged based on the execution ID 511 in ascending order or in descending order.

[0044] In Figure 6, screen 601 includes detection date and time 611, location information 612 in the image storage section 212 of the storage folder to which the inspection image is linked, algorithm 613 used to detect the abnormality, inspection image 631, wide-area detection results 632, linear detection results 633, and settings 621 for switching between displaying and hiding the detection results 632 and 633.

[0045] The detection result 632 and the detection result 633 are superimposed on the inspection image 631. The number of detection results displayed on the inspection image 631 may be one or more. In this case, the order of superimposition of the detection results superimposed on the inspection image 631 is determined in advance so that the user can clearly visually recognize each abnormality detection result. For example, a water leak is represented as a wide-range detection result 632, whereas a crack is represented as a detection result 633 based on the length and width of a line segment. Therefore, the image analysis result management unit 215 superimposes the water leak detection result 632 on the inspection image 631, and further superimposes the crack detection result 633 on the detection result 632. This allows the user to clearly visually recognize all the detection results 632 and the detection result 633 on the inspection image 631.

[0046] The user sets the acquisition conditions (output conditions) for the detection result image on a setting screen described later. The image processing device 101 can provide the detection result image based on the set acquisition conditions to the user via the user terminal 102. Note that the image processing device 101 transmits only the detection result image to the user terminal 102, but is not limited to this, and may transmit a set of the anomaly detection result and the inspection image, or only the anomaly detection result to the user terminal 102.

[0047] <Workflow for obtaining the anomaly detection results for the structure folder> The following describes a process of detecting an abnormality in an image (that is, an inspection image) of a structure and then providing a detection result image to the user terminal 102 (user's local PC).

[0048] <Save the inspection image> 7 is a diagram illustrating a hierarchical structure of a storage folder for inspection images according to the first embodiment.

[0049] A hierarchical structure 701 of storage folders for inspection images 731a to 731g is made up of a structure folder 711 and structure part folders 721a to 721c. For example, inspection image 731a can be stored in either the structure folder 711 or the part folder 721a.

[0050] The structure folder 711 is, for example, a folder representing the name of a bridge, and is illustrated as “XX bridge.” The structure folder 711 stores, for example, an inspection image 731g.

[0051] Part folders 721a to 721c are folders representing, for example, the names of parts that make up a bridge, and are illustrated as "A pier," "B pier," and "C pier." Part folder 721a stores inspection images 731a to 731c. Part folder 721b stores inspection image 731d. Part folder 721c stores inspection images 731e to 731f.

[0052] The inspection object in this embodiment is, for example, a bridge pier having a rectangular prism shape. The user inspects each of the four sides of one bridge pier for abnormalities such as cracks and / or water leakage. A case where the user photographs the inspection object (for example, a bridge pier) on-site will be described. Each of the four sides of a bridge pier has a vast area, so it is difficult to capture the entire range of the inspection object in a single image with sufficient image resolution to detect abnormalities. Therefore, the user repeatedly photographs a portion of each inspection object (bridge pier) in close-up while gradually moving the shooting range.

[0053] Then, the user performs image processing such as enlargement, reduction, rotation, projective transformation, color adjustment, and noise removal on the captured images. After that, the user joins the images that have been subjected to image processing to generate one composite image, thereby obtaining inspection images 731a to 731g (corresponding to images 1 to 7). The user inspects the four sides of the inspection object (pier). The four inspection images (not shown) are then linked to the structure folder 711 and the structure parts folder 721 and stored in the image storage unit 212. For example, when detecting anomalies in three piers that constitute a bridge, the user creates a "bridge" folder and creates three "pier" folders under the "bridge" folder. The user then stores the inspection images of parts of the pier in the three "pier" folders.

[0054] <Perform anomaly detection> FIG. 8 is a diagram illustrating the relationship between an inspection image and an execution ID according to the first embodiment.

[0055] The image analysis unit 213 performs anomaly detection processing on the inspection images 731a to 731g in the image storage unit 212 to detect cracks, water leaks, and the like. When the image analysis unit 213 performs anomaly detection processing on one inspection image, one or more types of anomaly are detected. The inspection images 731a to 731g are managed in association with the execution IDs of the execution information 811. The execution information 811 includes a plurality of execution IDs 1 to 6 that indicate what anomaly detection processing is to be performed on which inspection image. For example, execution ID 1 indicates that a "crack" is to be detected in the inspection image 731c (shown in image 3).

[0056] For example, inspection image 731a (illustrated in image 1) is managed in association with execution ID 3. In this case, one inspection image and one execution ID are associated in a one-to-one relationship. Alternatively, one inspection image and multiple execution IDs may be associated in a one-to-many relationship. For example, inspection image 731c (illustrated in image 3) is managed in association with two execution IDs (illustrated in execution ID 1 and execution ID 3).

[0057] <Management of image analysis results> The image analysis result management unit 215 displays a list of the anomaly detection results stored in the image analysis result storage unit 214 and displays details. When anomaly detection is performed multiple times, the anomaly detection results for multiple times are displayed on the screen 501 of FIG. 5. The screen 601 of FIG. 6 displays the anomaly detection results and the anomaly detection result images associated with one execution ID. The user can obtain the detection result images associated with one execution ID by selecting the result acquisition button 514 on the screen 501. At this time, when the user selects the batch result acquisition button 515, all the detection result images associated with all the inspection images on which the image analysis (anomaly detection) process was performed are obtained. In addition, when the user selects the result detail display button 516 to check the detection results in more detail, the image processing device 101 displays the screen 601. In addition, when the user wants to check the detailed detection results on the screen 501, the result detail display button 516 associated with the execution ID 511 is pressed. The image processing device 101 displays a screen 601 on the user terminal 102. This allows the user to check a detection date and time 611, an algorithm 613 (for example, a crack / leak detection model), a detection result 632, a detection result 633, and an inspection image 631 via the screen 601.

[0058] <Acquisition of images of abnormality detection results> FIG. 9 is a schematic diagram of an inspection report according to the first embodiment.

[0059] The inspection report 901 is a document submitted to the person requesting the inspection after inspecting a structure for abnormalities. The inspection report 901 includes a detection result image 911 and abnormality detection results 921. The user needs the following detection result images to attach to the inspection report 901. For example, the detection result images include one image that displays all abnormality detection results, and multiple images that display only the abnormality detection results for each abnormality type.

[0060] Also, to streamline the creation work of the inspection report 901, there is a need to acquire all detection result images in the structure folder at once. In order to acquire detection result images including desired abnormality detection results, the user needs to set acquisition conditions for the detection result images. Specific examples of detection result images including desired abnormality detection results will be described with reference to Figs. 10 to 12.

[0061] FIG. 10 is a diagram showing a detection result image displaying a deformation detection result for each deformation type according to the first embodiment.

[0062] When the user sets the abnormality detection results for each abnormality type to be displayed in the inspection image, the image analysis result acquisition unit 216 displays a detection result image 1001 in Fig. 10(a) and a detection result image 1002 in Fig. 10(b) on the user terminal 102. Fig. 10(a) shows a detection result image 1001 that displays the abnormality detection result of only "cracks (illustrated by linear areas)". Fig. 10(b) shows a detection result image 1002 that displays only "water leakage (illustrated by polygonal areas)".

[0063] FIG. 11 is a diagram showing a detection result image displaying all abnormality detection results according to the first embodiment.

[0064] When the user sets to display all abnormality detection results in one inspection image, the image analysis result acquisition unit 216 displays a detection result image 1101 in Fig. 11. In the detection result image 1101, the detection results of "cracks (illustrated by linear areas)" and "water leakage (illustrated by polygonal areas)" are displayed. In this way, the detection result image 1101 is the same as the image in which the detection result image 1001 and the detection result image 1002 are superimposed and displayed.

[0065] FIG. 12 is a diagram illustrating the hierarchical structure of folders in a zip file according to the first embodiment.

[0066] When the user sets to acquire the abnormality detection results in the structure folder all at once, the image analysis result acquisition unit 216 transmits a compressed file in zip format shown in FIG. 12 to the user terminal 102. The zip file can combine multiple files into one file while maintaining the hierarchical structure of the storage folder for the inspection images in FIG. 7, thereby improving user convenience. The structure folder 1211 (illustrated as "XX bridge.zip") stores part folders 1221a to 1221c that store the inspection images of each part of the structure. The part folder 1221a stores the detection result image 1231a (illustrated as images 1 to 3). The part folder 1221b stores the detection result image 1231b (illustrated as image 4). The part folder 1221c stores the detection result image 1231c (illustrated as images 5 and 6).

[0067] The image analysis result acquisition unit 216 presents the setting screen shown in FIG. 13 or FIG. 14 to the user in order to transmit the detection result image 1001 and the detection result image 1002 in FIG. 10, or the detection result image 1101 in FIG.

[0068] When the user sets the acquisition conditions for the detection result image on the setting screen of Fig. 13 or 14 via the user terminal 102, the setting information is transmitted to the image processing device 101. Thereafter, the image processing device 101 generates a detection result image based on the setting conditions (output conditions) and transmits the detection result image to the user terminal 102. This enables the user to obtain from the image processing device 101 a detection result image that displays the abnormality detection result set by the user.

[0069] Fig. 13 is a diagram illustrating an acquisition setting screen for anomaly detection results according to the first embodiment. A user sets whether or not to acquire anomaly detection results linked to one execution ID for each type of anomaly via a screen 1301 in Fig. 13. Various screens in this specification, such as the screen 1301 and a screen 1401 described later, are user interfaces that allow a user to set acquisition of anomaly detection results. The screens 1301 and 1401 are displayed on a display device of the user terminal 102 or the like.

[0070] The screen 1301 includes a display 1311, a setting 1312, an execution button 1321, and a cancel button 1322. The display 1311 displays the execution ID to be downloaded. For example, the display 1311 displays "Downloading the results of execution ID xx" as a message to be displayed on the user terminal 102. The setting 1312 indicates the download conditions (corresponding to the first output conditions) for the anomaly detection results. The setting 1312 displays, for example, "Separate the detection result images by anomaly type." If the check box for the setting 1312 is checked, this indicates that the user has selected the download conditions for the anomaly detection results. On the other hand, if the check box for the setting 1312 is not checked, this indicates that the user has not selected the download conditions for the anomaly detection results.

[0071] When the user presses (selects) the execute button 1321, the image processing device 101 generates a detection result image based on the settings 1312. The image processing device 101 transmits the generated detection result image to the user terminal 102, and the download of the detection result image is completed. When the user selects the cancel button 1322, the download of the detection result image to the user terminal 102 is stopped.

[0072] A method for acquiring a detection result image using the setting screen of FIG. 13 will be described with reference to FIG.

[0073] In FIG. 8, the inspection image 731b (illustrated as "Image 2") is associated with the execution ID 2. Here, the execution ID 2 is an ID for detecting "cracks and water leakage" in the image 2. When the user sets the screen 1301 in FIG. 13 to acquire the detection result of "cracks" and the detection result of "water leakage" in the image 2 (i.e., when the setting 1312 is selected), the following process is performed. That is, the image analysis result acquisition unit 216 transmits the detection result image 1001 and the detection result image 1002 to the user terminal 102. On the other hand, when the user does not set the screen 1301 to acquire the detection result of "cracks" and the detection result of "water leakage" in the image 2 (i.e., when the setting 1312 is not selected), the following process is performed. That is, the image analysis result acquisition unit 216 transmits the detection result image 1101 to the user terminal 102.

[0074] Fig. 14 is a diagram illustrating a collective acquisition setting screen for abnormality detection results according to the first embodiment. On a screen 1401 in Fig. 14, the user sets whether or not to collectively acquire abnormality detection results of structures.

[0075] The screen 1401 includes settings 1411, a structure folder 1421, structure part folders 1422a to 1422c, execution IDs 1431a to 1431f, an execution button 1441, and a cancel button 1442.

[0076] The setting 1411 has a function of setting a condition (corresponding to the second output condition) for downloading all the abnormality detection results of the structure folder at once, and displays, for example, "obtain all the detection result images of the structure folder at once." The structure folder 1421 stores the parts folders 1422a to 1422c of the structure, and is illustrated as "XX bridge." The parts folders 1422a to 1422c store the inspection images of each part of the structure, and are illustrated as "A pier" to "C pier." The execution IDs 1431a to 1431f are buttons for selecting the execution IDs associated with the images 1 to 6. For example, the image 3 is associated with two execution IDs (execution ID 1 and execution ID 3) as described in FIG. 8. The user can change the execution ID corresponding to the image 3 by selecting an arbitrary execution ID from the pull-down menu of the execution ID 1431c.

[0077] When the user selects execute button 1441, image analysis result acquisition unit 216 acquires detection result images based on settings 1411 and execution IDs 1431a to 1431f from image analysis result storage unit 214. Image analysis result acquisition unit 216 displays the acquired detection result images on user terminal 102, and downloading of the detection result images is completed. On the other hand, when the user selects cancel button 1442, downloading of the detection result images is stopped.

[0078] When the user presses the get result button 514 on screen 501 to obtain the detection result image to be used in the inspection report 901, the image analysis result obtaining unit 216 displays the screen 1301 of Fig. 13 on the user terminal 102. On the other hand, when the user presses the get all results button 515 on screen 501, the image analysis result obtaining unit 216 displays the screen 1401 of Fig. 14 on the user terminal 102.

[0079] First, the operation of the image processing device 101 based on the settings of the screen 1301 by the user will be described.

[0080] When the user selects the result acquisition button 514 on the screen 501, the image analysis result acquisition unit 216 displays a screen 1301 on the user terminal 102. On the screen 1301, the user selects setting 1312 for acquisition conditions of the detection result image to set that the detection result image is to be acquired for each deformation type.

[0081] When the user selects the execute button 1321 with the setting 1312 selected, the image analysis result acquisition unit 216 classifies the anomaly detection results by anomaly type. Then, the image analysis result acquisition unit 216 generates the detection result image 1001 and the detection result image 1002 in Fig. 10 by superimposing each of the anomaly detection results for each anomaly type on the inspection image. Finally, the image analysis result acquisition unit 216 transmits a compressed file in zip format or the like that stores the detection result image 1001 and the detection result image 1002 to the user terminal 102.

[0082] The detection result image 1001 and the detection result image 1002 may be stored in a compressed file in a compression format other than the zip format. If the user selects the execute button 1321 without selecting the setting 1312, the image analysis result acquisition unit 216 transmits the detection result image 1101 in FIG.

[0083] Next, the operation of the image processing device 101 based on the user's settings on the screen 1401 will be described.

[0084] The user can also collectively download detection result images linked to a structure folder. When the user presses the collective result acquisition button 515 on the screen 501, the image analysis result acquisition unit 216 transmits a screen 1401 to the user terminal 102. On the screen 1401, the user selects a setting 1411 for collectively downloading the abnormality detection results linked to the structure folder. Then, the user selects the execution ID 1431 linked to the abnormality detection result that the user wants to acquire for each inspection image 1423.

[0085] In the execution ID 1431, an execution ID associated with each inspection image is displayed based on information held by the image analysis result storage unit 214. The user sets an execution ID 1431a to 1431f for each inspection image via the screen 1401, and selects an execution button 1441 to instruct acquisition of a detection result image. The image analysis result acquisition unit 216 acquires the detection result image associated with the execution ID set by the user from the image analysis result storage unit 214. The image analysis result acquisition unit 216 constructs the structure folder 1211 of FIG. 12 in the same manner as the hierarchical structure 701 of the storage folder of the inspection image in FIG. 7. Then, the image analysis result acquisition unit 216 stores the detection result image in each part folder of the structure folder 1211, and transmits a compressed file including the structure folder 1211 to the user terminal 102. The structure folder 1211 including the detection result image may be compressed in zip format or by another compression method.

[0086] At least one of the setting 1312 on the screen 1301 and the setting 1411 for downloading all of the abnormality detection results on the screen 1401 at once may be displayed on the user terminal 102. Furthermore, the abnormality detection result is not limited to a detection result image in which the abnormality detection result is superimposed on the inspection image, and may be only the abnormality detection result data before being superimposed on the inspection image.

[0087] <Flowchart explaining how to obtain a detection result image associated with one execution ID> Fig. 15 is a flowchart for explaining a process in which the image processing device according to the first embodiment transmits a detection result image associated with one execution ID. The process in Fig. 15 is realized by the CPU of the control unit 111 of the image processing device 101 executing a program in the ROM of the non-volatile memory 113.

[0088] In S1501, the image analysis result acquisition unit 216 of the image processing device 101 displays a screen 501 on the user terminal 102. When the image analysis result acquisition unit 216 receives a user selection of a result acquisition button 514 associated with one execution ID 511, the process proceeds to S1502. Here, the user selection of the result acquisition button 514 means an instruction to acquire a detection result image.

[0089] In S1502, the image analysis result acquisition unit 216 displays the screen 1301 on the user terminal 102. The image analysis result acquisition unit 216 determines whether or not to acquire a detection result image for each deformation type, based on the user selection of the setting 1312 and / or the execute button 1321. Here, if the image analysis result acquisition unit 216 determines that the setting 1312 and the execute button 1321 have been selected by the user (Yes in S1502), the process proceeds to S1503. On the other hand, if the image analysis result acquisition unit 216 determines that the setting 1312 has not been selected by the user and that the execute button 1321 has been selected by the user (No in S1502), the process proceeds to S1504.

[0090] In S1503, the image analysis result acquisition unit 216 acquires the anomaly detection results of the inspection image associated with the execution ID 511 received in S1501 from the image analysis result storage unit 214 of the image processing device 101. Then, the image analysis result acquisition unit 216 divides the acquired anomaly detection results by anomaly type. The image analysis result acquisition unit 216 generates a detection result image for each anomaly type by superimposing each of the divided anomaly detection results on the inspection image, and proceeds to processing in S1505. For example, the detection result images for each anomaly type are detection result image 1001 and detection result image 1002 (see FIG. 10).

[0091] In S1504, the image analysis result acquisition unit 216 acquires the anomaly detection results of the inspection image associated with the execution ID 511 received in S1501 from the image analysis result storage unit 214 of the image processing device 101. Then, the image analysis result acquisition unit 216 generates a detection result image including all the anomaly detection results by superimposing all the anomaly detection results on the inspection image, and proceeds to processing in S1505. For example, the detection result image including all the anomaly detection results is the detection result image 1101 (see FIG. 11).

[0092] In S1505, the image analysis result acquisition unit 216 stores the detection result image generated in S1503 or S1504 in the structure folder 1211 or the part folder specified by the execution ID 511. Then, the image analysis result acquisition unit 216 collects the structure folder 1211 or the part folder into a zip file, transmits the zip file to the user terminal 102, and ends the process.

[0093] <Flowchart explaining how to collectively acquire all detection result images linked to a structure folder> 16 is a flowchart for explaining the process of the image processing device according to the first embodiment transmitting all detection result images in a structure folder at once. The process in FIG. 16 is realized by the CPU of the control unit 111 of the image processing device 101 executing a program in the ROM of the non-volatile memory 113.

[0094] In S1601, the image analysis result acquisition unit 216 of the image processing device 101 displays a screen 501 on the user terminal 102. When the image analysis result acquisition unit 216 receives a user selection of the batch result acquisition button 515, the process proceeds to S1602. Here, the user selection of the batch result acquisition button 515 means an instruction to acquire all detection result images in the structure folder.

[0095] In S1602, the image analysis result acquisition unit 216 displays the screen 1401 on the user terminal 102. The image analysis result acquisition unit 216 accepts an instruction as to whether or not to collectively download all detection result images in the structure folder based on the user selection of the setting 1411 and / or the execute button 1441. If the image analysis result acquisition unit 216 accepts the user selection of the setting 1411 and the execute button 1441 (i.e., if an instruction to collectively download all detection result images in the structure folder is accepted) (Yes in S1602), the image analysis result acquisition unit 216 advances the process to S1603. If the setting 1411 is not selected by the user and the execute button 1441 is selected by the user (i.e., if an instruction to collectively download all detection result images in the structure folder is not accepted) (No in S1602), the image analysis result acquisition unit 216 advances the process to S1605.

[0096] In S1603, the user sets execution IDs 1431a to 1431f for each inspection image 1423 on the screen 1401, and selects the execute button 1441. The image analysis result acquisition unit 216 accepts the execution IDs 1431a to 1431f set by the user for each inspection image 1423, and proceeds to the process in S1604.

[0097] In S1604, the image analysis result acquisition unit 216 acquires the anomaly detection results of each inspection image 1423 associated with each execution ID 1431a to execution ID 1431f set in S1603 from the image analysis result storage unit 214. Then, the image analysis result acquisition unit 216 generates each detection result image by superimposing each anomaly detection result on each inspection image 1423, and proceeds to the process at S1606.

[0098] In S1605, the image analysis result acquisition unit 216 identifies one execution ID linked to the batch result acquisition button 515 selected by the user in S1601. Then, the image analysis result acquisition unit 216 acquires anomaly detection results of one or more inspection images linked to the identified one execution ID from the image analysis result storage unit 214. Here, the identified one execution ID may be associated with multiple inspection images. Then, the image analysis result acquisition unit 216 generates one or more detection result images by superimposing one or more anomaly detection results on one or more inspection images, and proceeds to processing in S1606.

[0099] In S1606, the image analysis result acquisition unit 216 stores the detection result images generated in S1604 or S1605 in each of the part folders 1221a to 1221c of the structure folder 1211. Then, the image analysis result acquisition unit 216 collects the structure folder 1211 into a zip file, transmits the zip file to the user terminal 102, and ends the process.

[0100] As described above, according to the first embodiment, the user can combine and use the detection result image acquisition setting for each deformation type and the batch acquisition setting for all detection result images in the structure folder. This allows the user to efficiently acquire detection result images to be used in the inspection report from among multiple detection result images.

[0101] In this embodiment, all detection result images in the structure folder are sent at once, but the present invention is not limited to this, and a specific detection result in the structure folder may be sent. In this case, in S1603, in addition to each execution ID that can be selected for each inspection image 1423, options such as "not to send" or "not to download" are prepared so that the user can select. This makes it possible to acquire detection result images more efficiently in cases where, for example, pier A and pier B are to be reported, but pier C is not to be reported.

[0102] Second embodiment In the first embodiment, the image analysis result acquisition unit 216 displayed the screen 1301 in FIG. 13 or the screen 1401 in FIG. 14 on the user terminal 102 in accordance with the user settings for the screen 501 in FIG. 5. Also, in the first embodiment, the acquisition settings for the detection result image were accepted from the user, and a compressed file including the detection result image based on the user's acquisition setting conditions (first output conditions or second output conditions) was transmitted to the user terminal 102. In the second embodiment, the screen 1701 in FIG. 17 is displayed on the user terminal 102 instead of the screen 501. In the second embodiment, the differences from the first embodiment will be described.

[0103] FIG. 17 is a diagram illustrating a screen for making settings for acquiring a detection result image according to the second embodiment.

[0104] The screen 1701 includes an execution ID 1711 , a detection execution date and time 1712 , a status 1713 indicating whether the detection execution was successful or unsuccessful, a result acquisition button 1714 for acquiring the detection results, and a result detail display button 1715 .

[0105] 5. Detection execution date and time 1712 has the same function as the detection date and time 512. Status 1713 has the same function as the status 513. Result details display button 1715 has the same function as the result details display button 516.

[0106] The get result button 1714 has the functions of both the get result button 514 for acquiring a detection result image linked to one execution ID 1711, and the get all result button 515 for acquiring all detection result images in the structure folder at once.

[0107] When the user selects the get result button 1714, both the screen 1301 and the screen 1401 may be displayed on the user terminal 102. Furthermore, the condition items set on the screen 1301 and the screen 1401 may be displayed on one screen by the user selecting the get result button 1714. A specific example will be described below.

[0108] <Flowchart for retrieving multiple anomaly detection results linked to a structure folder by anomaly type> Fig. 18 is a flowchart for explaining that the image processing device according to the second embodiment executes a process relating to a combination of the process in Fig. 15 and the process in Fig. 16. The process in Fig. 18 is realized by the CPU of the control unit 111 of the image processing device 101 executing a program in the ROM of the non-volatile memory 113.

[0109] In S1801, the image analysis result acquisition unit 216 displays the screen 1701 on the user terminal 102. The image analysis result acquisition unit 216 accepts a user selection of the result acquisition button 1714 associated with a specific execution ID, and advances the process to S1802.

[0110] In S1802, the image analysis result acquisition unit 216 displays the screen 1301 on the user terminal 102. The image analysis result acquisition unit 216 determines whether or not to acquire a detection result image for each deformation type based on the user selection of the setting 1312 and / or the execute button 1321. If the image analysis result acquisition unit 216 determines that a detection result image for each deformation type is to be acquired (Yes in S1802), the process proceeds to S1803. On the other hand, if the image analysis result acquisition unit 216 determines that a detection result image for each deformation type is not to be acquired (No in S1802), the process proceeds to S1804.

[0111] In S1803, the image analysis result acquisition unit 216 acquires the anomaly detection results of the inspection image linked to the execution ID selected by the user in S1801 from the image analysis result storage unit 214. Then, the image analysis result acquisition unit 216 divides the acquired anomaly detection results by anomaly type. The image analysis result acquisition unit 216 generates a detection result image for each anomaly type by superimposing each of the divided anomaly detection results on the inspection image, and proceeds to processing in S1805.

[0112] In S1804, the image analysis result acquisition unit 216 acquires the anomaly detection results of the inspection image associated with the execution ID selected by the user in S1801 from the image analysis result storage unit 214. Then, the image analysis result acquisition unit 216 generates a detection result image by superimposing all of the acquired anomaly detection results on the inspection image, and proceeds to processing in S1805.

[0113] In S1805, the image analysis result acquisition unit 216 displays the screen 1401 on the user terminal 102. Then, the image analysis result acquisition unit 216 receives an instruction as to whether or not to collectively download all of the detection result images in the structure folder, based on the user selection of the setting 1411 and / or the execute button 1441. If the image analysis result acquisition unit 216 receives an instruction to collectively download the detection result images (Yes in S1805), the process proceeds to S1806, and if the instruction is not received (No in S1805), the process proceeds to S1808.

[0114] In S1806, the user sets execution IDs 1431a to 1431f for each inspection image 1423 on the screen 1401, and selects the execute button 1441. The image analysis result acquisition unit 216 accepts the execution IDs 1431a to 1431f set by the user for each inspection image 1423, and proceeds to the process in S1807.

[0115] In S1807, the image analysis result acquisition unit 216 acquires the anomaly detection results of each inspection image 1423 associated with each execution ID 1431a to execution ID 1431f set in S1805 from the image analysis result storage unit 214. Then, the image analysis result acquisition unit 216 generates each detection result image by superimposing each anomaly detection result on each inspection image 1423, and proceeds to the process at S1809.

[0116] In S1808, the image analysis result acquisition unit 216 identifies one execution ID linked to the result acquisition button 1714 selected by the user in S1801. Then, the image analysis result acquisition unit 216 acquires anomaly detection results of one or more inspection images linked to the identified execution ID from the image analysis result storage unit 214. Here, the identified execution ID may be associated with multiple inspection images. Then, the image analysis result acquisition unit 216 generates each detection result image by superimposing each anomaly detection result on one or more inspection images, and proceeds to processing in S1809.

[0117] In S1809, the image analysis result acquisition unit 216 stores the detection result images generated in S1807 or S1808 in each of the part folders 1221a to 1221c of the structure folder 1211. Then, the image analysis result acquisition unit 216 collects the structure folder 1211 into a zip file, transmits the zip file to the user terminal 102, and ends the process.

[0118] The first series of processes from S1802 to S1804 (processing in FIG. 15) is executed before the second series of processes from S1805 to S1808 (processing in FIG. 16), but this is not limited thereto. The second series of processes may be executed before the first series of processes.

[0119] Also, for example, on one screen 1401, settings 1312, settings 1411, structure folder 1421, part folder 1422, inspection image 1423, and execution IDs 1431a to 1431f may be displayed.

[0120] Third embodiment In the third embodiment, a method in which a user sets a deformation type for each part folder of a structure will be described. The image analysis result acquisition unit 216 transmits (outputs) all detection result images of the structure folder corresponding to the deformation type set by the user for each part folder to the user terminal 102 in a batch.

[0121] FIG. 19 is a diagram illustrating a screen for setting acquisition conditions for a detection result image for each part folder of a structure according to the third embodiment.

[0122] The screen 1901 includes a setting 1911 , a selection 1921 , an execute button 1931 , and a cancel button 1932 .

[0123] The setting 1911 has a setting function for acquiring all detection result images of the structure folder and a function for setting a deformation type (i.e., execution ID) for each part folder of the structure, and is illustrated by, for example, two check boxes. To the right of the upper check box of the setting 1911, "Acquire all detection result images of the structure folder". On the other hand, to the right of the lower check box of the setting 1911, "Specify deformation type for each part folder" is displayed. The selection 1921 has a function for setting a deformation type for each part folder of the structure, and is displayed as a pull-down menu. The pull-down menu can include, for example, "All", "Crack", and "Water Leak" as shown in FIG. 19, but may also include a deformation type set by the user. When the execute button 1931 is selected by the user, downloading of the detection result images based on the setting 1911 and the selection 1921 is started. When the cancel button 1932 is selected by the user, downloading of the detection result images is not executed.

[0124] The user selects two check boxes for Settings 1911 on screen 1901, selects the desired deformation type from Selection 1921 provided to the right of each component (here, Pier A, Pier B, and Pier C), and selects Execute button 1931. This allows the user to acquire all of the detection result images displaying the deformation of the deformation type selected in the pull-down menu of Selection 1921. For example, the user can acquire a detection result image of Pier A displaying "all" deformations, a detection result image of Pier B displaying "cracks," and a detection result image of Pier C displaying "leakage." In this case, "all" deformations include "cracks" and "leakage."

[0125] The advantages of acquiring detection result images using the above method are as follows. For example, when a user wants to attach detection result images showing deformations of some deformation types to an inspection report 901, the detection result images for the inspection report 901 can be acquired efficiently using the method of the third embodiment. Alternatively, when a user wants to display deformations of different deformation types in each detection result image for each component folder of a structure and check the display results, the detection result images can be acquired and checked efficiently using the method of the third embodiment. As described above, it is possible to acquire detection result images for each component folder showing deformations of the deformation types desired by the user, so that the time required for the user to create an inspection report can be reduced compared to the conventional method.

[0126] (Fourth embodiment) In the first and second embodiments, a method for downloading a detection result image for each type of abnormality by a user's settings on the screen 1301 has been described. However, there are cases where a user wants to attach a detection result image such as the one below to the inspection report 901. For example, the detection result image is an image that displays the detection result of a abnormality width / area (e.g., crack width, water leakage range) that is equal to or greater than a certain level. Therefore, in the fourth embodiment, a method for acquiring a detection result image having a abnormality width and area that is equal to or greater than a certain level on a screen 2001, which will be described later, will be described.

[0127] FIG. 20 is a diagram illustrating a setting screen for acquiring a detection result image having a deformation width and a deformation area equal to or larger than a certain level according to the fourth embodiment.

[0128] The screen 2001 includes a setting 2011 and an input section 2012. The setting 2011 has a function of specifying the amount and area of ​​the abnormality to be displayed in the detection result image (corresponding to the third output condition), and is displayed with a check box. The input section 2012 is provided so that the user can input specific numerical values ​​of, for example, the crack width and the water leakage range as information on the abnormality to be displayed in the detection result image.

[0129] The user selects the setting 2011 check box on the screen 2001, and the input function of deformation information is enabled. Then, the user sets information on the deformation width and deformation area for each of the component folders of the structure by selecting the input section 2012. Alternatively, the deformation width and deformation area may be set collectively for all the structure folders. Moreover, the deformation width and deformation area may be set by the user selecting an option provided in advance in the input section 2012, or may be set by the user directly inputting a desired numerical value into the input section 2012.

[0130] Furthermore, the fourth embodiment can include, for example, a form for setting the type of deformation, the area of ​​deformation, and the width of deformation of the detection result image to be downloaded by using it in combination with the first to third embodiments, which allows the user to more easily obtain the detection result image that the user desires.

[0131] Fifth embodiment For example, a detection result image that does not have a specific abnormality area does not include information that the person requesting the inspection wants to confirm, and therefore does not need to be attached to the inspection report 901. Therefore, in the fifth embodiment, a setting (hereinafter also referred to as a filtering setting) for excluding detection result images in which no abnormality of a specific abnormality type was detected as an output target to the user terminal 102 will be described.

[0132] 21 is a diagram illustrating a filtering setting screen for detection result images according to the fifth embodiment. The filtering setting refers to a setting for excluding, from the targets to be output to the user terminal 102, detection result images in which no abnormality of a specific abnormality type was detected or in which no abnormality was detected.

[0133] The screen 2101 includes a selection section 2111. The selection section 2111 allows the user to perform the following settings (corresponding to the fourth output condition) for all the inspection images in the structure folder 1211. There are two options under "crack" in the selection section 2111: "Include all" and "Download only images with detection". If the user selects "Include all", the image analysis result acquisition section 216 stores the detection result images of detected and undetected "cracks" in the parts folder of the structure folder 1211. On the other hand, if the user selects "Download only images with detection", the image analysis result acquisition section 216 stores only the detection result images of detected "cracks" in the parts folder of the structure folder 1211. Similarly, there are two options under "water leakage" in the selection section 2111: "Include all" and "Download only images with detection". When the user selects either "Include all" or "Download only images with detection", the processing of the image analysis result acquisition unit 216 is as explained in the example of "cracks", so a detailed explanation will be omitted. Furthermore, a case where "Include all" and "Do not download" are present as two options under "No abnormality detected" in the selection unit 2111 will be explained. When the user selects "Include all", the image analysis result acquisition unit 216 stores the detection result images in which no abnormality has been detected in the parts folder of the structure folder 1211. On the other hand, when the user selects "Do not download", the image analysis result acquisition unit 216 does not store the detection result images in which no abnormality has been detected in the parts folder of the structure folder 1211.

[0134] In this way, the user selects an option from the selection unit 2111 provided for each of the items "crack," "water leakage," and "no abnormality detected" on the screen 2101. The image analysis result acquisition unit 216 generates a detection result image by superimposing the abnormality detection result based on the setting of the selection unit 2111 on the inspection image, and transmits a compressed file including the generated detection result image to the user terminal 102. By acquiring the detection result image in this manner, the user can acquire the detection result image required to be attached to the inspection report 901 with higher accuracy than before. This makes it possible to prevent the downloading of unnecessary detection result images that do not include a specific abnormality or that include almost no abnormality, thereby shortening the user's work time.

[0135] The fifth embodiment may be implemented in combination with the first to fourth embodiments.

[0136] (Other embodiments) In one embodiment, in the process of collectively acquiring detection result images of structures, the user designates an execution ID for each inspection image (i.e., selects the detection result image to be acquired), but this is not limited to the above. For example, the detection result image to be acquired may be automatically selected by the image processing device 101 based on the latest abnormality detection result executed using each inspection image.

[0137] In one embodiment, if the user does not set not to acquire detection result images of no abnormality detected on the detection result image acquisition setting screen, the detection result images of no abnormality detected are transmitted to the user terminal 102, but this is not limited thereto. For example, if no abnormality of any abnormality type is included in the detection result image, the image processing device 101 may automatically execute a process of excluding the above detection result images from the acquisition targets.

[0138] In one embodiment, a method of acquiring only an anomaly detection result for each anomaly type or a detection result image when acquiring a detection result image has been described, but this is not limited to this. For example, metadata such as the image resolution and detection date and time of the inspection image may be acquired at the same time when acquiring a detection result image.

[0139] In the embodiment, the method of acquiring the detection result image as is has been described, but the present invention is not limited to this. For example, if the size of the detection result image is equal to or larger than a certain size, the detection result image may be compressed before being acquired.

[0140] In one embodiment, the image analysis result acquisition unit 216 divides the anomaly detection results or the detection result images for each anomaly type when acquiring the detection result images, but this is not limited to this. For example, after dividing the detection result images for each anomaly type, the divided detection result images are stored in part folders for each anomaly type. After that, a compressed file storing a structure folder including part folders for each anomaly type may be acquired.

[0141] In one embodiment, no restrictions are placed on the structure folders that the image analysis result acquisition unit 216 can acquire when acquiring the detection result image, but this is not limited thereto. For example, the image processing device 101 automatically determines whether or not the structure folder is to be output based on the information of the structure folder. After that, a process may be automatically performed to not acquire the structure folder that is determined not to be output. For example, when detecting an abnormality in an inspection image, the image processing device 101 attaches information (e.g., a mark) indicating that the analysis is for testing to the detection result image and registers the information in the software. Thereby, when acquiring the detection result image, it is also possible to automatically exclude the execution result having the registration information of the test analysis from the acquisition target.

[0142] In the embodiment, the image processing device 101 does not have a function for editing the abnormality detection result, but the present invention is not limited to this. For example, if the image processing device 101 has a function for editing the abnormality detection result, the image processing device 101 may perform a process of acquiring both the data of the detection result before and after editing when acquiring the image analysis result.

[0143] In the embodiment, when the user sets all the detection results in the structure folder to be downloaded at once, the execution ID for each inspection image is selected on the screen, but this is not limited to the above. For example, an interface may be provided that displays the detection result images on the screen and allows the user to select the detection result image on the screen.

[0144] The "dimension information" described in the first embodiment is not limited to the planar rectangular coordinate system, but may be information based on the latitude and longitude coordinate system or other coordinate systems.

[0145] In one embodiment, a zip file is used when providing the detection result image to the user terminal 102, but this is not limiting. For example, other compression formats that can combine (compress) multiple files into one while maintaining the hierarchical structure of the storage folder for the inspection images may be used instead of the zip format.

[0146] (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.

[0147] The disclosure of this specification includes the following image processing device, image processing system, method, and program. (Item 1) A storage means for storing a first detection result including first deformation information detected in a first image of a first part folder of a subject, and a second detection result including second deformation information detected in a second image of a second part folder different from the first part folder; A receiving means for receiving at least one of a first output condition for outputting the first detection result and the second detection result for each type of anomaly associated with the first image and the second image, and a second output condition for outputting at least one of the first detection result and the second detection result; and an output means for outputting a file including at least one of the first part folder including at least a part of the first detection result and the second part folder including at least a part of the second detection result based on a result of reception by the reception means. 13. An image processing device comprising: (Item 2) the receiving means further receives a third output condition for outputting at least a part of each of the first detection result and the second detection result based on settings of an area and a width of a deformation associated with each of the first image and the second image, or associated with each of the first part folder and the second part folder. 2. The image processing device according to item 1, (Item 3) The receiving means further receives a selection of a specific type of anomaly from the types of anomalies associated with each of the first image and the second image. 3. The image processing device according to item 1 or 2. (Item 4) The receiving means further receives a fourth output condition that does not output at least one of the first detection result and the second detection result when at least one of the first detection result and the second detection result does not include a predetermined type of abnormality, or when at least one of the first detection result and the second detection result does not include an abnormality. 4. The image processing device according to any one of items 1 to 3, (Item 5) The receiving means further receives a selection of a specific type of defect from among the types of defects associated with the first part folder and the second part folder, respectively. 5. The image processing device according to any one of items 1 to 4, (Item 6) The output means outputs the file including one or more of the first part folders and the second part folders classified by the type of the deformation. 6. The image processing device according to any one of items 1 to 5, (Item 7) The output means outputs the file compressed in a predetermined format. 7. The image processing device according to any one of items 1 to 6, (Item 8) At least one of the first part folder and the second part folder, and the first detection result and the second detection result has a name based on information on the type of the deformation. 8. The image processing device according to any one of items 1 to 7, (Item 9) The first detection result includes at least one of the first image displaying the first deformation information and the first deformation information, The second detection result includes at least one of the second image displaying the second deformation information and the second deformation information, 9. The image processing device according to any one of items 1 to 8, (Item 10) The first deformation information and the second deformation information include at least cracks and water leakage, 10. The image processing device according to any one of items 1 to 9, (Item 11) the receiving means further includes presenting a user interface including at least one of the first output condition and the second output condition. 11. The image processing device according to any one of items 1 to 10, (Item 12) An image processing device according to any one of items 1 to 11, A user terminal that communicates with the image processing device. 1. An image processing system comprising: (Item 13) a storage step of storing a first detection result including first deformation information detected in a first image of a first part folder of a subject, and a second detection result including second deformation information detected in a second image of a second part folder different from the first part folder; A receiving process for receiving at least one of a first output condition for outputting the first detection result and the second detection result for each type of anomaly associated with the first image and the second image, and a second output condition for outputting at least one of the first detection result and the second detection result; and an output step of outputting a file including at least one of the first part folder including at least a part of the first detection result and the second part folder including at least a part of the second detection result based on a reception result of the reception step. A method comprising: (Item 14) 12. A program for causing a computer to function as the image processing device according to any one of items 1 to 11.

[0148] 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]

[0149] 101 Image processing device 111 Control section 112 Volatile Memory 113 Non-volatile memory 114 Storage Devices 115 Input Device 116 Output Device 117 Communication Equipment 118 System Bus 211 Image Management Department 212 Image Storage Unit 213 Image Analysis Unit 214 Image analysis results storage section 215 Image analysis results management department 216 Image analysis result acquisition unit 221 System Bus

Claims

1. A storage means for storing a first detection result, which includes first deformation information representing deformation of the structure detected in a first image stored in a first folder corresponding to a first part of the structure, among a plurality of folders provided to correspond to each of a plurality of parts constituting the structure that is the subject of the object, and a second detection result, which includes second deformation information representing deformation of the structure detected in a second image stored in a second folder corresponding to a second part of the structure different from the first part, A receiving means that accepts the selection of either a first output condition for outputting each of the first detection result and the second detection result according to the type of deformation, or a second output condition for outputting at least one of the first detection result and the second detection result without separating them by type of deformation, If the first output condition is selected, the output means outputs a file containing a folder storing the first detection results for each type of deformation indicated by the first deformation information, and a folder storing the second detection results for each type of deformation indicated by the second deformation information, and if the second output condition is selected, the output means outputs a file containing the detection results targeted for output by the second output condition stored in the folder corresponding to the detection results. An image processing apparatus characterized by having

2. The receiving means further receives a third output condition that limits the output target of the first detection result and the second detection result based on the first output condition or the second output condition, based on the deformation area and width settings set for each of the first image and the second image, or for each of the first folder and the second folder. The image processing apparatus according to feature 1.

3. The receiving means further accepts the selection of the type of deformation to be output from among the types of deformation detected in the first image and the second image, respectively. The image processing apparatus according to feature 1.

4. The receiving means further accepts a fourth output condition for each of the first and second detection results, which excludes the detection result from output if the detection result does not include a predetermined type of deformation, or if the detection result does not include any deformation. The image processing apparatus according to feature 1.

5. The receiving means further accepts the selection of the type of deformation to be output from among the types of deformation detected in the images stored in the first and second folders, respectively. The image processing apparatus according to feature 1.

6. When the first output condition is selected, the output means classifies the first detection result and the second detection result according to the type of deformation, and outputs a file containing the classified detection results in folders according to the type of deformation. The image processing apparatus according to feature 1.

7. The file output from the output means is a compressed file that maintains a hierarchical structure in which the first folder and the second folder are located within a structure folder corresponding to the structure. The image processing apparatus according to feature 1.

8. At least one of the first folder, the second folder, the first detection result, and the second detection result has a name based on information of the corresponding type of abnormality. The image processing apparatus according to feature 1.

9. The first detection result includes at least one of the first image on which the first deformation information is displayed, and the first deformation information. The second detection result includes at least one of the second image on which the second deformation information is displayed, and the second deformation information. The image processing apparatus according to feature 1.

10. The first detection result includes a first detection result image in which the first deformation information is superimposed on the first image, and the second detection result includes a second detection result image in which the second deformation information is superimposed on the second image. The image processing apparatus according to feature 1.

11. The output means stores metadata in the file, along with the first detection result and the second detection result, indicating at least one of the image resolution of the first image and the second image and the detection date and time. The image processing apparatus according to feature 1.

12. When the first output condition is selected, the output means places type folders, each set out for each type of deformation, into the first folder and the second folder, respectively, and outputs a file containing detection results that include deformation information of the type corresponding to the type folder to the type folder. The image processing apparatus according to feature 1.

13. A storage step of storing in a storage unit a first detection result including first deformation information representing deformation of the structure detected in a first image stored in a first folder corresponding to a first part of the structure, among a plurality of folders provided to correspond to each of a plurality of parts constituting the structure which is the subject of the data, and a second detection result including second deformation information representing deformation of the structure detected in a second image stored in a second folder corresponding to a second part of the structure which is different from the first part, A receiving step that accepts the selection of either a first output condition that outputs each of the first detection result and the second detection result according to the type of deformation, or a second output condition that outputs at least one of the first detection result and the second detection result without separating them by type of deformation. If the first output condition is selected, the output process outputs a file containing folders for each type of first detection result, separated by the type of deformation indicated by the first deformation information, and folders for each type of second detection result, separated by the type of deformation indicated by the second deformation information. If the second output condition is selected, the output process outputs a file containing folders for each detection result, in which the detection results targeted for output by the second output condition are stored in the folders corresponding to those detection results. An image processing method characterized by including

14. A program for causing a computer to function as an image processing device according to any one of claims 1 to 12.

Images