Information processing apparatus, information processing method, and program

Abstract

To detect recurrence of deformations.SOLUTION: An information processing apparatus selects a second deformation corresponding to a first deformation of a structure in a first image, from deformations of the structure in a second image picked up chronologically subsequent to the first image. The information processing apparatus estimates a first repaired area indicating a new repaired portion of the structure in the second image with respect to the time when the first image is picked up. The information processing apparatus determines whether a portion where the first deformation and the second deformation match each other is present in the first repaired area. The information processing apparatus determines whether the deformations of the structure recur on the basis of the determination as to whether the matched portion is present in the first repaired area.SELECTED DRAWING: Figure 6

Description

Technical Field

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

Background Art

[0002] Infrastructure such as the walls of bridges and tunnels on highways is used while being repaired every few decades. Therefore, the soundness of the members of the structure is regularly evaluated, and a determination is made as to whether repair is necessary. However, in recent years, due to the aging of infrastructure and a shortage of personnel, labor-saving in operations has been demanded, and deformation detection by video analysis has begun to be put into practical use.

[0003] In the inspection of infrastructure structures using images, deformation may be detected from images of the structure walls taken at different times, and the degree of progress of each deformation may be calculated by obtaining the difference between the detected deformations. Patent Document 1 discloses a technique of searching for similar past cracks from the center of gravity of the feature points of cracks and calculating the difference in the width of the maximum crack. Patent Document 2 discloses a technique of associating the deformations in images taken at two times and determining the state of their secular changes.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] When repair is performed, it is assumed that the abnormalities detected in the previous inspection have disappeared or decreased in the next inspection. However, there are cases where, despite the repair, the root cause has not been addressed and the abnormalities recur. For example, if the abnormality is due to pressure imbalance in the infrastructure structure, or inflow of anti-freezing water or rainwater on the road surface, or an abnormality in a location that is susceptible to wind or waves and is easily damaged, it is considered highly likely that recurrence will occur after repair. In the technology described in the above-mentioned patent document, although it is possible to extract the degree of progression of abnormalities in the previous and subsequent inspections, there is a problem that it is difficult to detect whether such abnormalities have recurred after repair.

[0006] An object of the present invention is to detect recurrence of abnormalities.

Means for Solving the Problems

[0007] In order to achieve the object of the present invention, for example, an information processing apparatus according to an embodiment includes the following configuration. That is, a selection means for selecting a second abnormality corresponding to a first abnormality of a structure in a first image from abnormalities of the structure in a second image captured later in time series than the first image, a first estimation means for estimating a first repair region indicating a new repair location of the structure in the second image with respect to the imaging time point of the first image, a first determination means for determining whether or not there is a matching portion between the first abnormality and the second abnormality in the first repair region, and a second determination means for determining whether or not the abnormality of the structure has recurred based on the determination of whether or not there is the matching portion in the first repair region.

Effects of the Invention

[0008] Detect recurrence of abnormalities.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the invention according to the claims. Although a plurality of features are described in the embodiments, not all of these plurality of features are essential for the invention, and the plurality of features may be arbitrarily combined. Further, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant descriptions are omitted.

[0011] [Embodiment 1] Basically, in the case of abnormalities such as cracks confirmed by inspection, changes in the direction of increase occur over time, such as the cracks extending or becoming thicker, efflorescence, water leakage, or the area of the abnormal part such as peeling becoming larger. However, when repairs are made to the abnormalities, the abnormalities detected during the regular inspection may decrease. For example, when repairs are made to a part determined to be severely damaged and in need of repair by the previous regular inspection using a construction method such as resin injection, the inspection result in the next inspection may be that the abnormality detected in the previous inspection does not exist or has decreased.

[0012] The information processing apparatus according to the present embodiment determines whether recurrence has occurred in the repaired portion of the abnormality. For this purpose, the information processing apparatus according to the present embodiment selects, from the abnormalities of the structure in the second image captured later in time series than the first image, a second abnormality corresponding to the first abnormality of the structure in the first image. Next, the information processing apparatus estimates a first repair area indicating a new repair portion of the structure in the second image with respect to the imaging time of the first image. Further, the information processing apparatus determines whether there is a matching portion between the first abnormality and the second abnormality in the first repair area, and determines whether the abnormality of the structure has recurred based on the determination of whether the matching portion exists in the first repair area.

[0013] Here, an example of a method of detecting the same abnormality existing in the area repaired between two time points using two images captured at different times and notifying the user will be described. In particular, the information processing apparatus according to the present embodiment can be used to perform so-called infrastructure inspection (inspection for whether there are abnormalities such as cracks on the surface of the structure) for determining the soundness of a structure such as a bridge and the degree of deterioration from the previous time.

[0014] Hereinafter, the terms used in this embodiment will be described. The "object to be inspected" according to this embodiment refers to a structure that is the object of the inspection process by the information processing apparatus. The "abnormality" according to this embodiment refers to an abnormality that occurs in the structure. For example, in a concrete structure, cracks, peeling or spalling of concrete, efflorescence, exposed rebar, rust, water leakage, dripping water, corrosion, damage (defect), cold joints, precipitates, or junk are regarded as abnormalities. In the following, cracks will basically be used as the abnormality for explanation, but different abnormalities may be used.

[0015] Also, the "repair location" according to this embodiment is a location where repair has been performed on the abnormality of the structure. The repair location here may be a location where resin has been filled in the crack site, a location where the surface has been painted, or a location where a sheet has been attached to the structure. Such repair of the abnormality can be arbitrarily performed according to the type of the abnormality or the conditions desired by the user.

[0016] The "first image" and the "second image" according to this embodiment are images of the same wall surface of the same structure taken at different times. In this embodiment, the second image is an image taken later in the time series than the first image.

[0017] Regarding the outline of the aging change determination that is the premise of this embodiment, an example of calculating the matching part and the degree of progress of general cracks will be used for explanation. The "aging change determination" according to this embodiment is a process of associating a first change state and a second change state, and obtaining the matching part and the difference between the associated change states. FIG. 1 shows, as an example of an image of an infrastructure, images 101 and 102 of the same wall surface of a bridge taken at different times. Image 101 is an image taken several years (for example, 5 years) before the imaging time of image 102. On the wall surface of the infrastructure, new cracks or an increase in crack width due to aging, or an expansion of the damaged area such as efflorescence, water leakage, or reinforcement exposure occurs. In image 101, cracks 103, 105 and water leakage 104 are imaged, and in image 102, cracks 106, 107 and water leakage 108 are imaged.

[0018] The crack 106 on image 102 is the result of the progression of the crack 103 on image 101, and the crack 107 is a newly generated crack in image 102. Also, the repair area 109 on image 102 is the repair location corresponding to the position of the crack 105. Here, the damage degree of the crack 105 is estimated, and repair has been performed because the crack 105 has become a repair target according to the estimated damage degree, and the repair location 109 has occurred. In the inspection process of the repair, due to such repair, a previously detected change state may not be detected, or the degree of the change state may be reduced.

[0019] The information processing apparatus according to the present embodiment stores, in association with each change detected from an image, width information indicating the width of the change. Table 110 in FIG. 1 is a diagram showing an example of such width information. Here, each crack is assigned a change ID, and width information is associated with each ID. Here, the width information is information indicating the maximum value of the crack width (for example, in terms of estimated distance, set distance, or the number of pixels in the image). Here, the change in the image is indicated by change region information indicating the coordinate position, shape, or width of the change, such as a crack or steel bar exposure. In the present embodiment, it is assumed that width information is generated by the user inputting the coordinate position, shape, or width of the change in the image in association with the change, but the change information is not particularly limited to being obtained in this way. For example, the information processing apparatus may identify the change in the image and obtain each piece of information by detecting the change from the image using a model pre-trained by machine learning.

[0020] The calculation of the progress degree of the change (crack) according to the present embodiment is performed by comparing corresponding changes (cracks) in two images. For example, the information processing apparatus can compare the change region information between images for the crack and obtain the matching part and the non-matching part (difference) between them. Note that the information processing apparatus may correct such misalignments while calculating the progress degree of the change, assuming that there are position misalignments or shape misalignments caused by differences in imaging conditions or detection conditions between crack data at different times.

[0021] [Overall Outline] FIG. 2 is a block diagram showing an example of the configuration of an image analysis system 200 (hereinafter, may be simply referred to as "system 200") including an information processing apparatus 202 according to the present embodiment. The system according to the present embodiment determines secular change regarding deformation of structures in two images, and determines whether or not the deformation has recurred. The system 200 according to the present embodiment can be used, for example, for infrastructure such as bridges, tunnels, or roads, to estimate deformation in two images captured periodically or irregularly, and to infer the degree of progress and recurrence of deformation in the repair area from the states of deformation at two points in time.

[0022] The image analysis system 200 includes a camera 201, an information processing apparatus 202, a mobile terminal 203, an image storage server 204, an image analysis server 205, and a result storage server 206. Further, the information processing apparatus 202, the mobile terminal 203, the image storage server 204, the image analysis server 205, and the result storage server 206 are communicably connected by the same network 207. The network 207 may be the Internet or a local network.

[0023] The camera 201 is an imaging device. The camera 201 according to the present embodiment transfers an image to be detected for deformation, which is captured by the user at an arbitrary timing, to the information processing apparatus 202 via a storage device such as a USB or via a network.

[0024] The information processing apparatus 202 transmits the image transferred from the camera 201 to the image storage server 204 or the image analysis server 205. Further, the information processing apparatus 202 performs each process related to the determination process of whether or not recurrence has occurred in the repair location of the deformation according to the present embodiment. Details of each process including the secular change analysis process performed by the information processing apparatus 202 will be described later with reference to FIG. 4. Further, the information processing apparatus 202 issues various instructions such as detection of deformation or secular change analysis to the image analysis server 205. Further, the information processing apparatus 202 issues various instructions such as acquisition of the analysis processing results managed by the image analysis server 205. Further, the information processing apparatus 202 notifies (for example, displays) the user of the acquired analysis results.

[0025] The mobile terminal 203 transmits the image to be detected for deformation captured by the camera 201 to the image storage server 204 or the image analysis server 205. Also, similar to the information processing device 202, the mobile terminal 203 performs detection of deformation, an instruction for aging change analysis, an instruction for obtaining the analysis processing result, and display of the obtained analysis result.

[0026] The image storage server 204 receives and stores the image to be analyzed transmitted from the information processing device 202 or the mobile terminal 203. The image analysis server 205 receives the image transmitted from the information processing device 202 or the mobile terminal 203.

[0027] Similar to the information processing device 202, the image analysis server 205 is capable of executing aging change analysis processing. In the present embodiment, it will be described that the information processing device 202 and the image analysis server 205 cooperate to execute aging change analysis processing, but one of these may execute all of the aging change analysis processing. For example, the image analysis server 205 can receive an aging change analysis instruction from the information processing device 202 or the mobile terminal 203 and acquire the first image specified from the image storage server 204 and the second image capturing the same wall surface of the same infrastructure together. Also, the image analysis server 205 can perform an analysis process of inferring the deformation of the wall surface using the acquired images, extracting the matching part and the difference in deformation between the first image and the second image, and transmitting the result to the result storage server 206. Further, the image analysis server 205 can extract the area that was not repaired in the first image but was repaired in the second image using the acquired images, determine whether the aforementioned matching part exists in the new repair area (described later), and transmit the determination result to the result storage server 206. Also, the image analysis server 205 may perform display of the analysis result instead of the information processing device 202 or the mobile terminal 203.

[0028] The result storage server 206 receives and stores the analysis results such as the deformation position information transmitted from the information processing device 202 or the image analysis server 205, the image with the deformation superimposed, the deformation difference information between the first image and the second image, and the matching part of the deformation in the new repair area.

[0029] Note that although each device including the camera 201, the information processing device 202, the mobile terminal 203, and the image analysis server 205 included in the system 200 according to this embodiment is a different device, it is not particularly limited in this way as long as it can execute similar processing. For example, the information processing device 202 may have an imaging function and function as the camera 201, or the information processing device 202 may be built into the mobile terminal 203. In this way, the information processing device 202 may have some or all of the functions of other devices included in the system 200. Further, although the information processing device 202 according to this embodiment is described as being a personal computer (PC), it may be a different device capable of executing similar processing, such as a server or a mobile terminal. Also, in this embodiment, the image to be processed is described as being acquired from the camera 201 (or the mobile terminal 203), but it may be acquired from a fixed-point camera (not shown) connected to the network 207, or may be acquired from an external server via the Internet.

[0030] Hereinafter, the configuration of the image analysis server 205 according to this embodiment and the processing executed by the image analysis server 205 will be described. Note that in this embodiment, the information processing device 202 can execute each process including the aging change analysis process, and the system may be configured such that the information processing device 202 executes some or all of the processes described below as being executed by the image analysis server 205.

[0031] [Hardware Configuration] FIG. 3 is an example of the hardware configuration of the image analysis server 205 according to the present embodiment. The image analysis server 205 includes a Central Processing Unit (CPU) 300, a Randam Access Memory (RAM) 301, a Read Only Memory (ROM) 302, a network interface (I / F) 303, a storage I / F 304, and a storage 305 connected to a bus 306.

[0032] The CPU 300 comprehensively controls each functional unit connected via the bus. The RAM 301 is used as a main memory of the CPU 300 that can be accessed at high speed and a temporary storage area such as a work memory. The ROM 302 stores a processing program shown in a flowchart described later, a device driver, and the like. The network I / F 303 is used when communicating with other devices. The storage I / F 304 inputs and outputs data to and from the storage 305. The storage 305 stores images received from the information processing device 202 or the mobile terminal 203, images received from the image storage server 204, analysis results obtained by analyzing the images, and programs and an OS operating inside the image analysis server 205, and information necessary at power-on is read into the RAM 301. An external storage device may be used to perform the same role as the storage 305. Here, the external storage device can be realized, for example, by a medium (storage medium) and an external storage drive for realizing access to the medium. As such a medium, for example, a flexible disk (FD), a CD-ROM, a DVD, a USB memory, an MO, or a flash memory can be used. Further, the external storage device may be a server device connected via a network.

[0033] Note that the information processing device 202, the image storage server 204, and the result storage server 206 are assumed to have the same hardware configuration as the image analysis server 205. Further, it can also be configured as an alternative to the hardware device by software that realizes functions equivalent to those of the above components.

[0034] [Functional Block Diagram] FIG. 4 is an example of a block diagram showing the functional configuration of the image analysis server 205. The image analysis server 205 includes, as functional blocks, a storage unit 400, a management unit 401, a communication unit 402, a deformation detection unit 403, a deformation comparison unit 404, a repair area estimation unit 405, a new repair estimation unit 406, and a recurrence deformation determination unit 407. Each of these functional units is realized by the CPU 300 expanding the program stored in the ROM 302 into the RAM 301 and executing processing according to each flowchart described later. Then, the image analysis server 205 holds the execution results of each process in the RAM 301. Further, for example, when configuring hardware as an alternative to software processing using the CPU 300, an arithmetic unit or a circuit corresponding to the processing of each functional unit described here may be configured.

[0035] The storage unit 400 stores images, metadata, plan view coordinate information, deformation detection results, repair area detection results, and aging change determination results used for aging change determination. The management unit 401 stores the images acquired through the communication unit 402 in the storage unit 400. The management unit 401 also manages the images captured at different times, the deformation area information corresponding to each image, or the structure information related to the structure stored in the storage unit 400. Each piece of information stored in the storage unit will be described later.

[0036] The communication unit 402 receives images and metadata in a communicable format from the information processing device 202, the mobile terminal 203, or the image storage server 204 via the network I / F 303 and outputs them to the management unit 401. The communication unit 402 also acquires various information such as deformations detected in the deformation comparison process described later, information on repair areas, aging change determination results, and coordinate information of deformations that have recurred in the new repair area, stores them in the result storage server 206, and causes them to be displayed on the information processing device 202 or the mobile terminal 203.

[0037] The image analysis server 205 performs aging change analysis processing on two or more input images. The aging change analysis processing according to the present embodiment compares, as the aging change analysis processing, the first deformation of the structure in the first image with the second deformation of the structure in the second image corresponding to the first deformation, and determines whether recurrence has occurred in the repair location of the deformation (recurrence determination). Hereinafter, such aging change analysis processing will be described.

[0038] The deformation detection unit 403 detects deformations in the image. In the present embodiment, the deformation detection unit 403 detects cracks from the first image and the second image and estimates their widths. Here, the detection of deformations may be performed by detecting a deformation region by image analysis processing using a machine learning model, or may be performed by acquiring deformation region information such as the above-described table 110 attached to each image. The deformation region detected by the image analysis processing shall be represented by linear or planar coordinate positions in the image coordinates and the plan view coordinates.

[0039] The deformation comparison unit 404 associates the deformations in the first image detected by the deformation detection unit 403 with the deformations in the second image. The association of deformations may be performed, for example, by evaluating the matching rate of each deformation in the first image with respect to each deformation in the second image (for example, using a machine learning model), and setting the deformation with the highest matching rate as the corresponding deformation. Also, for example, the center positions of the deformations (for example, the center points at both ends) may be calculated respectively, and the deformations with the closest center position distance (or within a predetermined distance) between the first image and the second image may be set as the corresponding deformations. When a positional deviation or a shape deviation caused by a difference in imaging conditions or a difference in detection conditions is detected, such a deviation may be corrected before the association of the deformations is performed. Hereinafter, although the processing will be described as being performed on a pair of deformations associated in this way, it is possible to perform the same processing on other pairs. The association of the first deformation and the second deformation and the alignment of the images can be performed using known techniques, and detailed description thereof will be omitted.

[0040] Next, the deformation comparison unit 404 extracts the matching portion between the first deformation and the second deformation. The image analysis server 205 can extract the matching portion between the first deformation and the second deformation based on, for example, the coordinate information included in the deformation region information. Also, for example, the image analysis server 205 may superimpose the deformation region information and extract the overlapping region as the matching portion. Here, in addition to the matching portion between the first deformation and the second deformation, the deformation comparison unit 404 extracts the portion of the second deformation that does not match the first deformation as the difference between those deformations.

[0041] In cases where the area of the corresponding second deformation is enlarged or the width of the crack is widened compared to the first deformation, it is considered that there is a positive difference. On the other hand, in cases where the area of the corresponding second deformation is reduced or the width of the crack is narrowed compared to the first deformation, it is considered that there is a negative difference. FIG. 5 is a diagram showing the matching portions and differences of the respective deformations between the first image 501 and the second image 502. The first image 501 includes deformations 503, 504, and 505, and the second image 502 includes deformations 506, 507, and 508. The deformations 503 and 506, and the deformations 504 and 508 are associated as pairs of deformations at two time points by the deformation comparison unit 404. The matching portion image 509 shows the matching portion of the deformations between the first image 501 and the second image 502, and includes matching portions 510 and 511. The difference image 512 shows the difference in deformations between the first image 501 and the second image 502, and includes differences 513, 514, 515, and 516. The differences 513 and 515 are differences due to the enlargement of the deformation due to aging, and the difference 514 is a difference due to the occurrence of a new deformation. On the other hand, the difference 516 is a difference indicating a deformation that was detected in the first image but not detected in the second image, and since it cannot occur in the normal time-series change in the infrastructure, it indicates that it is the result of repair by repair.

[0042] The new repair estimation unit 406 estimates a first repair region (new repair region) indicating a new repair location of the structure in the second image with respect to the imaging time point of the first image. For example, the repair region estimation unit 405 can estimate, based on a portion of the first deformation that does not match the second deformation (a location where the first deformation exists but the second deformation does not exist) among the corresponding first deformation and second deformation (e.g., its peripheral region), as the new repair region. For example, the repair region estimation unit 405 may estimate a rectangular region including a portion of the first deformation that does not match the second deformation, or a region with a predetermined width added to that portion, as the new repair region.

[0043] Also, the new repair estimation unit 406 may extract a partial region that does not correspond to the repair region (second repair region) indicating the repair location of the structure in the first image from the repair region (third repair region) indicating the repair location of the structure in the second image, and use it as the new repair region. By extracting the repair region that exists only in the third repair region among the second repair region and the third repair region, it is possible to extract, as the new repair region, a region where it is considered that new repair has been performed during the period from imaging the first image to imaging the second image (during the periodic inspection).

[0044] For this purpose, the repair region estimation unit 405 can estimate a second repair region indicating the repair location of the structure in the first image and a third repair region indicating the repair location of the structure in the second image. The repair region estimation unit 405 according to the present embodiment can extract a repair region from the input image using, for example, a machine learning model. Here, the machine learning model can be trained by a known method such as semantic segmentation using learning data based on the difference in pixel values (RGB difference or shadow) from the surrounding region in the structure, shape, or chalk marks indicating the repair location, and detailed description thereof is omitted.

[0045] The recurrence deformation determination unit 407 determines whether or not there is a matching portion between the first deformation and the second deformation in the new repair area. As such determination, the recurrence deformation determination unit 407 may determine whether or not the matching portion extracted by the deformation comparison unit exists in the new repair area, or may determine whether or not there is a matching portion with the first deformation for the second deformation existing within the new repair area. Based on such determination results, the recurrence deformation determination unit 407 determines whether or not the deformation of the structure has recurred. When there is a matching portion in the new repair portion, the recurrence deformation determination unit 407 can determine that the deformation has recurred because it is considered that the same deformation has occurred in the area repaired after the previous inspection.

[0046] The process performed by the recurrence deformation determination unit 407 according to this embodiment will be described with reference to FIGS. 6(a) to 6(c). FIG. 6(a) shows an image 601 that is the first image and an image 602 that is the second image. The deformation 603 indicates the area of a crack detected in the first image, and the deformation 604 indicates the area of exposed reinforcement detected in the first image. In this case, an operation of repairing the structure for the deformation 604 is assumed between the imaging time of the image 601 and the imaging time of the image 602. The deformation 605 is a deformation of the second image, the difference 606 is the difference between the deformation 603 and the deformation 605, and the repair area 607 indicates the repaired area. In this case, since the deformation 605 does not exist inside the repair area 607, it is determined that the deformation has not occurred in the repair area (the deformation has not recurred). Thus, when the second partial area does not exist in the new repair area, the recurrence deformation determination unit 407 can determine that the recurrence of the deformation has not occurred. According to such determination, it becomes possible to present to the user as a recurrence determination result the difference 606 of the deformation existing outside the new repair area such as the deformation 605 in FIG. 6(a).

[0047] Figure 6(b) shows image 611 which is the first image and image 612 which is the second image. The deformation 613 indicates the area of water leakage detected in the first image. Also, the deformation 614 is a crack in the second image, and the area 615 indicates the repaired area repaired for the deformation 613. In this case, although the deformation 614 exists inside the area 615, since no deformation equivalent to the deformation 614 has occurred in the first image, it is determined that it is a newly occurred deformation in the repaired area (the deformation has not recurred). Thus, when there is a second sub-region in the new repair area but no matching part, the recurrence determination unit 407 can determine that the recurrence of the deformation has not occurred and a new deformation has occurred.

[0048] An example where a determination different from that of Fig. 6(b) is made is shown in Fig. 6(c). Fig. 6(c) shows image 621 which is the first image and image 622 which is the second image. The deformation 623 indicates the area of the crack detected in the first image. Also, the deformation 624 indicates the area of the crack detected in the second image, the matching part 625 is the matching part between the deformation 623 and the deformation 624, and the area 626 indicates the repaired area repaired for the deformation 623. In this case, since both the deformation 624 and the matching part 625 exist inside the area 626, it is presumed that the same deformation has occurred (the deformation has recurred) between the imaging time of the image 621 and the imaging time of the image 622 despite the repair of the target deformation. Thus, when there is a matching part in the new repair area, the recurrence determination unit 407 can determine that the recurrence of the deformation has occurred. According to such a determination, by determining whether the deformation occurring in the repair area matches the past one, it becomes possible to determine whether a new deformation has occurred at the repair location as in Fig. 6(b) or whether a past deformation has recurred as in Fig. 6(c).

[0049] [Explanation of the relationship between the image and the deformation area information and the structural information] In explaining the processing according to this embodiment, the relationship between the image and the deformation area information, and the structure information will be described. Here, in image inspection, the image obtained by imaging the wall surface of a structure is managed in association with a plan view. Fig. 7(a) shows an image 700 obtained by imaging the wall surface of a bridge attached to a plan view as an example of an infrastructure structure. In this plan view, there is a plan view coordinate system 702 with the point 701 at the upper left corner as the origin. Each "position of the image" is defined by the vertex coordinates at the upper left of the image in the plan view coordinate system 702. For example, when four divided images are taken for a plan view of a certain floor slab, the position of each image is the position of the vertex at the upper left of each image in the plan view coordinate system 702. The image is stored in the storage unit 400 together with the coordinate information indicating the position of the image.

[0050] For example, the images used in the image inspection of infrastructure structures may be large in size because they are taken at a high resolution (e.g., 1 mm per pixel) so as to confirm fine cracks and the like. For example, the image 700 attached to the plan view in Fig. 7(a) is an image of the floor slab of a 20 m × 10 m bridge. When the image resolution is 1.0 mm per pixel (1.0 mm / pixel), the image size of the floor slab of a 20 m × 10 m bridge is 20,000 pixels × 10,000 pixels.

[0051] The deformation area information according to this embodiment is information recording the result of (automatic) detection of deformations such as cracks occurring on a concrete wall surface, or the input result by a human. In explaining the processing according to this embodiment, it is assumed that the deformation area information is managed in association with the plan view. Fig. 7(b) shows a state 710 where the deformation area information corresponding to the image 700 attached to the plan view is attached to the plan view at the same position as the position in the image 700. The position of each deformation area information on the plan view is defined by the plan view coordinates of the pixels constituting the deformation area information.

[0052] Note that the deformed region information may be represented by vector data such as a polyline or a curve composed of a plurality of points. When the deformed region information is represented by vector data, the data volume is reduced and the representation becomes more concise. As an example of deformed region information other than cracks, there is steel bar exposure. When representing steel bar exposure by a polyline, it becomes a deformation having a region surrounded by the polyline. The attribute information included in the deformed region information is not limited to numerical information such as coordinates, width, length, and area, and other attribute information may be included.

[0053] The structure information according to this embodiment is information related to the structure of the inspection target, and includes various information such as the type or basic structure of the structure, various dimensions of the structure, member information, or the completion year. Further, the structure information may include, as repair records, information related to the maintenance of the structure such as the repair year, repair location, or repair method. In this embodiment, the structure information related to a specific position of the structure such as member information or repair records is stored together with the plan view coordinates which are the position information on the plan view. That is, it is assumed that the position of each member on the plan view or the position of the repair location on the plan view is stored as a part of the structure information. Therefore, through the plan view, the correspondence relationship between the structure information, the image, and the deformed region information can be obtained. The structure information is stored in the storage unit 400 together with the image and the deformed region information, and can be acquired by the management unit 401. Note that the information included in the structure information is not limited to the above information, and other information may be held. Also, as the structure information, information limited for each type may be held according to the type of the structure.

[0054] [Recurrence determination process according to Embodiment 1] FIG. 8 is a flowchart showing an example of the determination of secular change and the recurrence determination process of changes in the repair area by the image analysis server 205 according to the present embodiment. Hereinafter, each step (step) will be described by attaching an S to the beginning of those reference numerals. In the following, the data in which a change is detected from the first image is referred to as first change area information, and the data in which a change is detected from the second image is referred to as second change area information. In the present embodiment, when the image analysis server 205 receives a user input for starting processing, the processing of the flowchart in FIG. 6 is started. Hereinafter, the recurrence determination process according to Embodiment 1 will be described with reference to the flowchart in FIG. 6.

[0055] In S800, the change detection unit 403 detects and extracts changes in the first image and the second image acquired via the communication unit 402 or acquired from the storage unit 400. The changes extracted in the first image and the second image are acquired as coordinate information. In the following, the information on the change extracted from the first image is referred to as first change area information, and the information on the change extracted from the second image is referred to as second change area information.

[0056] In S801, the change comparison unit 404 associates the change in the first change area information with the change in the second change area information. Here, the change comparison unit 404 extracts and associates, with respect to the change in the first change area information, a change in the second change area information that exists at a plan view coordinate approximate to the plan view coordinate of the change in the first change area information (for example, exists within a predetermined distance). When there are a plurality of second changes at approximate coordinates for one first change, the similarity based on the shape of the change, the direction of the vector, or the like may be calculated, and the association with the second change having the highest similarity may be performed.

[0057] In S802, the change comparison unit 404 calculates the matching part and the difference of the change area from the paired first change area information and second change area information extracted in S801. In the example of FIG. 5, differences 513, 514, 515, and 516 are detected as differences in secular change in S802.

[0058] In S803, the repair area estimation unit 405 estimates the repair area in the first image and the repair area in the second image. In S804, the new repair estimation unit 406 estimates a new repair area indicating a new repair location of the structure in the second image with respect to the imaging time of the first image. Here, the new repair estimation unit 406 extracts, from the repair area of the first image and the repair area of the second image extracted in S803, the area determined to be the repair area only in the second image as the new repair area. Also, without performing S803, the peripheral area of the deformation detected in the first image that has disappeared in the second image may be estimated as the new repair area in S804.

[0059] In S805, the recurrence deformation determination unit 407 determines whether or not the second deformation detected in S800 exists within the new repair area estimated in S804. When the recurrence deformation determination unit 407 determines that the second deformation exists within the new repair area, the process proceeds to S807, and otherwise the process proceeds to S806.

[0060] In S806, the recurrence deformation determination unit 407 stores, in the storage unit 400, the matching part and the difference calculated in S802 for the second deformation determined to exist outside the new repair area in S805 as information indicating deformation due to aging, and ends the process. Also, the recurrence deformation determination unit 407 can notify the user of the information indicating deformation due to aging by a notification unit (not shown) of the information processing device 202 or the mobile terminal 203 (for example, via the communication unit 402). According to such processing, it becomes possible to present to the user the difference 606 of the deformation existing outside the new repair area such as the deformation 605 in FIG. 6(a) as the determination result of aging.

[0061] In S807, the recurrence deformation determination unit 407 determines whether or not there is a matching part with the first deformation for the second deformation determined to exist within the new repair area in S805. When it is determined that there is a matching part, the process proceeds to S809, and otherwise the process proceeds to S808.

[0062] In S808, as shown in the deformation 614 of FIG. 6(b), the management unit 401 does not associate with the first deformation, and stores the deformation area information of the second deformation determined in S805 to exist in the new repair area in the storage unit 400 as new deformation information in the repair area, and ends the process. Further, the management unit 401 can notify the user of the occurrence of a new deformation in the repair area through a notification unit (not shown) of the information processing device 202 or the mobile terminal 203 (for example, via the communication unit 402).

[0063] In S809, as shown in the matching part 625 of FIG. 6(c), the management unit 401 stores the matching part of the first deformation area information and the second deformation area information determined to exist in the new repair area in S807 in the storage unit 400 as the deformation information that has recurred in the repair area, and ends the process. Further, the management unit 401 can notify the user of the occurrence of the deformation that has recurred in the repair area through a notification unit (not shown) of the information processing device 202 or the mobile terminal 203 (for example, via the communication unit 402).

[0064] According to such determination, by determining whether the deformation that has occurred in the repair area matches the past one, it is possible to determine whether a new deformation has occurred at the repair location as shown in FIG. 6(b) or whether the past deformation has recurred as shown in FIG. 6(c).

[0065] FIG. 9 is an example of a UI displayed when performing aging change determination and recurrence determination on the information processing apparatus 202 or the mobile terminal 203 in the present embodiment. The screen 900 in FIG. 9(a) is an example of a screen for executing aging change determination and recurrence determination for the first image and the second image. Information 901 indicates the imaging date and image name of the first image, and information 902 indicates the imaging date and image name of the second image. Also, in FIG. 9(a), image 903 is shown as the first image, and image 904 is shown as the second image. Region 906 is a repair region estimated to have been newly repaired between 2017 and 2022, and display 905 is an emphasized display for making it easy to notice that a change has been detected in region 906, which is the newly repaired region. Notification 907 is a notification displayed as an alert when a change is detected in the newly repaired region. When notification 907 is pressed by the user, the management unit 401 causes the screen 900 to transition to the screen 910 in FIG. 9(b).

[0066] The screen 910 in Fig. 9(b) is an example of a repaired area detection change list screen that displays a list of changes detected in the repaired area that was the source of the alert by the notification 907. The list 911 shows a list of the cropped locations of the changes detected in the repaired area. The thumbnail 912 shows the thumbnail of an individual change in the list 911. When the thumbnail 912 is pressed by the user, the management unit 401 displays the change images 913 and 914 in the center of the screen. The change image 913 shows the first change of the selected thumbnail. The change image 914 shows the second change of the selected thumbnail. The user can refer to this screen to check the repaired locations and the occurred changes in the second image, and also compare the second change with the first image. Also, when the same change recurs despite the repair, by accepting the pressing of the button 915 by the user, the management unit 401 can register that the change has recurred. The registered recurring change may be displayed so that it can be confirmed on the screen that it has recurred, and mainly for the purpose of the inspection report, the user may be able to download information indicating such a recurring change together with the cropped change image or the detected change information. Also, the management unit 401 may similarly perform labeling, addition to the download target, etc. regarding the occurrence of new changes that occur only in the second image in the new repaired area.

[0067] According to such a configuration, it becomes possible to determine whether or not the change in the structure has recurred based on the determination of whether or not the matching part of the change between the first image and the second image exists in the new repaired area. Also, it becomes possible to detect the occurrence of new changes in the new repaired area. By notifying the user of such a recurrence determination result, it is possible to prevent cases where the user does not notice the occurrence of changes in the repaired area and the recurrence of past changes visually, and by being able to obtain the information necessary for the inspection report, it is possible to contribute to the efficiency of the inspection and the prevention of overlooking.

[0068] [Embodiment 2] The image analysis server (information processing apparatus) according to Embodiment 1 estimated a repair area from each image, and performed recurrence determination processing using a new repair area estimated based on the estimated repair area. On the other hand, the image analysis server according to Embodiment 2 acquired the repair area of each image based on user input, and estimated a new repair area based on the repair area thus acquired. The system according to the present embodiment has the same configuration as that shown in FIG. 2 of Embodiment 1 except that the image analysis server 1000 is provided instead of the image analysis server 205, and can execute the same processing, so overlapping explanations are omitted.

[0069] FIG. 10 is a block diagram showing an example of the functional configuration of the image analysis server 1000 according to the present embodiment. The image analysis server 1000 does not have a repair area estimation unit 405 and a new repair estimation unit 406, and instead of the storage unit 400, the communication unit 402, and the recurrence state determination unit 407, it includes a storage unit 1001, a communication unit 1002, and a recurrence state determination unit 1003, and has a repair area registration unit 1004. Since it has the same configuration as the image analysis server 205 of Embodiment 1, overlapping explanations are omitted.

[0070] The image analysis server 1000 according to the present embodiment acquires information on the repair area in each image based on an input by the user (for example, by accepting the upload of the coordinate information of the repair area). Here, the communication unit 1002 can acquire information on the repair area in each image by acquiring the content of the user input to an input unit (not shown) of the image analysis server 1000 or to the mobile terminal 203. The storage unit 1001 stores information on the repair area based on such user input.

[0071] The repair area registration unit 1004 stores information on the repair area based on such user input in the storage unit 1001. FIG. 11 is a diagram showing an example of the repair area according to the present embodiment and a file input by the user when registering the repair area.

[0072] FIG. 11(a) is a diagram showing an example of an image capturing a deformation being repaired based on first deformation region information. Deformation 1101 indicates a first deformation that is a crack, and region 1102 indicates a region where the crack is filled using a repair material. In the example of FIG. 11(a), the polyline region 1102 that has actually been repaired for deformation 1101 is used as the repair region, but the repair region is not necessarily limited to such a shape. For example, the repair region may be a wider rectangular region surrounding the repaired region 1102. FIG. 11(b) is a diagram showing an example of repair region registration information representing the repair region shown in FIG. 11(a) on plan view coordinates. By acquiring such a file as the repair region, the ID and area of the repair region, as well as the number of vertices of the polyline that is the repair region and the plan view coordinates of each vertex, can be stored as a repair history before aging change determination.

[0073] Based on such information of the repair region, the repair region registration unit 1004 estimates a new repair region. The estimation of the new repair region can be performed in the same manner as the new repair estimation unit 406 of Embodiment 1 based on the acquired information of the first deformation and the second deformation. Alternatively, a new repair region may be directly input by the user, and the repair region registration unit 1004 may set the new repair region by referring to the information. The repair region registration unit 1004 can include the information of the specified new repair region in the information of the repair region and store it in the storage unit 1001.

[0074] The recurrence deformation determination unit 407 executes the determination processes of S805 and S807 based on the information of the repair region stored in the storage unit 400. In this case, the recurrence deformation determination unit 407 aligns the positions of the first deformation, the second deformation, and the repair region based on the plan view coordinates, and determines whether there is an overlapping portion between the first deformation and the second deformation in the new correction region.

[0075] According to such processing, by acquiring from the user a file indicating the correction region, particularly a file corresponding to the plan view coordinates from the repair record before aging change comparison, it is possible to more accurately determine the occurrence of deformation and the overlapping determination between the overlapping portion and the deformation in the repair region.

[0076] [Embodiment 3] When the matching part of the first deformation and the second deformation does not exist in the new repair area, the image analysis server according to Embodiment 1 determines that the deformation has not recurred. However, it is conceivable that the deformation progresses to another type of deformation over time, for example, peeling or falling off from a crack and then progressing to reinforcement exposure. Therefore, in an actual structure, even if reinforcement exposure occurred at the time of imaging the first image and repair was carried out, and no recurrence of the same type of deformation occurred at the time of imaging the second image, the root cause may not have been addressed, and cases where different deformations such as cracks or peeling occur can be assumed.

[0077] From such a perspective, when the matching part of the first deformation and the second deformation does not exist in the new repair area, the image analysis server according to Embodiment 3 further determines that the deformation has recurred if a deformation similar to the second deformation occurs at the same position as the second deformation in the first image. According to such processing, it is possible to determine that a recurrence has occurred even for different deformations, considering cases where it is insufficient to only detect a recurrence of the same deformation in the repair area. The system according to the present embodiment has the same configuration as that shown in FIG. 2 of Embodiment 1 and can execute the same processing, so duplicate explanations are omitted.

[0078] The recurrence determination unit 407 according to the present embodiment searches for the first deformation corresponding to (existing in the vicinity of) the second deformation determined to exist within the new repair area. The first deformation searched for here may be the first deformation associated as a pair with the second deformation in Embodiment 1, or may be a different first deformation. Next, when it is determined that the first deformation and the second deformation existing in the vicinity as a result of such a search are similar, the recurrence determination unit 407 determines that the deformation has recurred.

[0079] Next, a case where it is determined that the first deformation and the second deformation are similar will be described. If the recurrence deformation determination unit 407 determines that both the first deformation and the second deformation, which are supposed to exist in the vicinity as a result of such a search, are deformations of a specific type, it can be determined that the deformation has recurred. Here, the specific type of deformation refers to a deformation where the occurrence location does not move, such as a crack, rust, peeling, detachment, or steel bar exposure. This is because efflorescence, water leakage, and rust juice are deformations that spread by riding on water, and even if they occur at the same position in the new repair area, the crack that is the water ejection point is not necessarily at the same position. Further, the recurrence deformation determination unit 407 may determine that the first deformation and the second deformation are similar when the second deformation is included in the first deformation supposed to exist in the vicinity. Further, the recurrence deformation determination unit 407 may determine that the first deformation and the second deformation are similar when the second deformation overlaps with the first deformation supposed to exist in the vicinity. Note that when a plurality of these conditions are satisfied, it may be determined that the first deformation and the second deformation are similar.

[0080] The process performed by the recurrence deformation determination unit 407 according to this embodiment will be described with reference to FIGS. 12(a) and 12(b). FIG. 12(a) shows an image 1201 that is the first image and an image 1202 that is the second image. The deformation 1203 indicates the area of steel bar exposure detected in the first image, and the deformation 1204 indicates the area of the detected crack that spreads around the steel bar exposure. Further, the deformation 1205 indicates the crack at the same position as the deformation 1204 in the second image, and the area 1206 indicates the repair area repaired for the deformation 1203. In this case, the deformation 1203 and the deformation 1205 are different deformations, and no difference is extracted in the aging change determination described in Embodiment 1. However, since it is assumed that the deformation 1205 will expand and eventually result in the recurrence of steel bar exposure like the deformation 1203 after peeling, the recurrence deformation determination unit 407 according to this embodiment determines that the deformation has recurred based on such a deformation.

[0081] In addition, FIG. 12(b) shows image 1211 which is the first image and image 1212 which is the second image. Deformation 1213 indicates the crack region detected in the first image. Also, deformation 1214 indicates the crack region of the second image, deformation 1215 indicates the water leakage region caused by deformation 1214, and region 1216 indicates the repair region repaired for deformation 1213. In this case, the deformation does not recur at the repaired location, but the upper 1214 is the problem and deformation 1215 has occurred. This deformation 1215 is a deformation that expands downward by water from the cause location, similar to efflorescence or rust juice. Therefore, deformation 1213 and deformation 1215 exist in the vicinity, and a part of deformation 1215 is included in region 1216. However, since deformation 1215 is not a specific type of deformation, it is not determined that the deformation in the repair region has recurred.

[0082] FIG. 13 is a flowchart showing an example of the aging change determination and the recurrence determination process of the deformation in the repair region by the image analysis server according to the present embodiment. The process shown in FIG. 13 is performed in the same manner as that in FIG. 8 except that S1301 is provided between S807 and S808, so overlapping explanations are omitted.

[0083] When it is determined in S807 that there is no matching part of the deformation in the new repair region, the recurrence deformation determination unit 407 advances the process to S1301. In S1301, the recurrence deformation determination unit 407 searches for the second deformation determined to exist in the new repair region and the first deformation existing in the vicinity of a certain value or less, and determines whether or not both such first deformation and second deformation are specific deformations. If both are specific deformations, the recurrence deformation determination unit 407 advances the process to S609, and if not, advances the process to S608. In this case, it is determined that the deformation in the repair region has recurred.

[0084] According to such a process, by determining whether or not there is a past deformation of a specific type in the vicinity of the deformation that has occurred in the repair region, it is possible to determine that recurrence has occurred even when the root cause has not been countered even if it is not the same deformation. Therefore, it is possible to more accurately determine the recurrence of the deformation in the repair region.

[0085] The disclosure of this specification includes the following information processing apparatus, information processing method, and program. (Item 1) Selection means for selecting, from the deformation of the structure in a second image captured later in time series than the first image, a second deformation corresponding to the first deformation of the structure in the first image; First estimation means for estimating a first repair region indicating a new repair location of the structure in the second image with respect to the imaging time point of the first image; First determination means for determining whether or not there is a matching portion between the first deformation and the second deformation in the first repair region; Second determination means for determining whether or not the deformation of the structure has recurred based on the determination of whether or not there is the matching portion in the first repair region; An information processing apparatus, comprising the above. (Item 2) The information processing apparatus according to Item 1, wherein the first estimation means estimates, as the first repair region, a partial region including a portion of the first deformation that does not match the second deformation. (Item 3) The information processing apparatus according to Item 1, further comprising second estimation means for estimating a first repair region indicating a repair location of the structure in the first image and a second repair region indicating a repair location of the structure in the second image, wherein the first estimation means estimates, as the first repair region, a partial region obtained by extracting a region that does not correspond to the first repair region from the second repair region. (Item 4) The information processing apparatus according to Item 3, wherein the second estimation means estimates the first repair region and the second repair region indicating the repair location of the structure based on a chalk mark indicating the repair location or a difference in pixel values between the structure and its surroundings. (Item 5) Further comprising acquisition means for acquiring a user input indicating a first repair area showing a repair location of a structure in the first image and a second repair area showing a repair location of the structure in the second image. The information processing apparatus according to item 1, wherein the first estimation means estimates, as the first repair area, a partial area obtained by extracting, from the second repair area, an area that does not correspond to the first repair area. (Item 6) The information processing apparatus according to any one of items 1 to 5, wherein the second determination means determines that the deformation has recurred when the matching portion exists in the first repair area. (Item 7) Further comprising third determination means for determining whether or not a second deformation in the first repair area is similar to a third deformation corresponding to the second deformation in the first image when the second deformation exists in the first repair area. The information processing apparatus according to item 6, wherein the second determination means further determines that the deformation has recurred when the matching portion does not exist in the first repair area and it is determined that the second deformation and the third deformation are similar. (Item 8) The information processing apparatus according to item 7, wherein the third determination means determines that the second deformation and the third deformation are similar when both the second deformation and the third deformation are deformations of a specific type. (Item 9) The information processing apparatus according to item 8, wherein the deformation of the specific type is a deformation in which the occurrence location of the deformation does not move. (Item 10) The information processing apparatus according to item 7, wherein the third determination means determines that the second deformation and the third deformation are similar when the third deformation is included in the second deformation or the third deformation overlaps with the second deformation. (Item 11) The second determination means further determines that a new change has occurred when the second change exists in the first repair area and the matching part does not exist in the first repair area, and is characterized in that, the information processing apparatus according to any one of items 1 to 10. (Item 12) The information processing apparatus further includes extraction means for extracting a matching part between the first change and the second change. Based on a determination as to whether or not the matching part extracted by the extraction means exists in the first repair area, the first determination means determines whether or not the change in the structure has recurred, and is characterized in that, the information processing apparatus according to any one of items 1 to 11. (Item 13) The information processing apparatus further includes notification control means for notifying the user of the recurrence of the change when it is determined that the change in the structure has recurred, and is characterized in that, the information processing apparatus according to any one of items 1 to 12. (Item 14) A step of selecting, from changes in the structure in a second image captured later in time series than the first image, a second change corresponding to the first change in the structure in the first image; A step of estimating a first repair area indicating a new repair location in the second image of the structure with respect to the imaging time point of the first image; A step of determining whether or not a matching part between the first change and the second change exists in the first repair area; Based on a determination as to whether or not the matching part exists in the first repair area, a step of determining whether or not the change in the structure has recurred; An information processing method, characterized by comprising the above steps. (Item 15) A program for causing a computer to function as each means of the information processing apparatus according to any one of items 1 to 13.

[0086] (Other embodiments) The present invention can also be realized by supplying a program that implements one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and causing one or more processors in a computer of the system or apparatus to read and execute the program. It can also be realized by a circuit (for example, ASIC) that implements one or more functions.

[0087] The invention is not limited to the above-described embodiments, and various changes and modifications can be made without departing from the spirit and scope of the invention. Therefore, the claims are appended to disclose the scope of the invention.

Description of Reference Numerals

[0088] 201: Camera, 202: Information Processing Apparatus, 203: Mobile Terminal, 204: Image Storage Server, 205: Image Analysis Server, 206: Result Storage Server

Claims

1. Selection means for selecting, from among deformations of a structure in a second image captured later in time series than the first image, a second deformation corresponding to a first deformation of the structure in the first image; First estimation means for estimating a first repair region indicating a new repair location of the structure in the second image with respect to the imaging time point of the first image; First determination means for determining whether or not a matching portion between the first deformation and the second deformation exists in the first repair region; Second determination means for determining whether or not the deformation of the structure has recurred based on the determination as to whether or not the matching portion exists in the first repair region; An information processing apparatus, comprising the above.

2. The information processing apparatus according to claim 1, wherein the first estimation means estimates, as the first repair region, a partial region including a portion of the first deformation that does not match the second deformation.

3. Further comprising second estimation means for estimating a first repair region indicating a repair location of the structure in the first image and estimating a second repair region indicating a repair location of the structure in the second image, The information processing apparatus according to claim 1, wherein the first estimation means estimates, as the first repair region, a partial region obtained by extracting a region in the second repair region that does not correspond to the first repair region.

4. The information processing apparatus according to claim 3, wherein the second estimation means estimates the first repair region and the second repair region indicating the repair location of the structure based on a chalk mark indicating the repair location or a difference in pixel values between the structure and its surroundings.

5. Further comprising acquisition means for acquiring a user input indicating the first repair region indicating the repair location of the structure in the first image and the second repair region indicating the repair location of the structure in the second image, The information processing apparatus according to claim 1, wherein the first estimation means estimates, as the first repair region, a partial region obtained by extracting a region in the second repair region that does not correspond to the first repair region.

6. The information processing apparatus according to claim 1, wherein the second determination means determines that the deformation has recurred when the matching portion exists in the first repair region.

7. When the second deformation exists in the first repair area, further comprising third determination means for determining whether the second deformation and a third deformation corresponding to the second deformation in the first image are similar; The information processing apparatus according to claim 6, wherein when the second determination means determines that the matching part does not exist in the first repair area and the second deformation and the third deformation are similar, the second determination means further determines that the deformation has recurred.

8. The information processing apparatus according to claim 7, wherein the third determination means determines that the second deformation and the third deformation are similar when both the second deformation and the third deformation are specific types of deformations.

9. The information processing apparatus according to claim 8, wherein the specific type of deformation is a deformation in which the occurrence location of the deformation does not move.

10. The information processing apparatus according to claim 7, wherein the third determination means determines that the second deformation and the third deformation are similar when the third deformation is included in the second deformation or the third deformation overlaps with the second deformation.

11. The information processing apparatus according to claim 1, wherein when the second deformation exists in the first repair area and the matching part does not exist in the first repair area, the second determination means further determines that a new deformation has occurred.

12. Further comprising extraction means for extracting a matching part between the first deformation and the second deformation; The information processing apparatus according to claim 1, wherein the first determination means determines whether the deformation of the structure has recurred based on a determination of whether the matching part extracted by the extraction means exists in the first repair area.

13. The information processing apparatus according to claim 1, further comprising notification control means for notifying the user of the recurrence of the deformation when it is determined that the deformation of the structure has recurred.

14. A step of selecting a second deformation corresponding to a first deformation of a structure in a first image from deformations of the structure in a second image captured later in time series than the first image; A step of estimating a first repair area indicating a new repair location of the structure in the second image with respect to the imaging time point of the first image; A step of determining whether or not there is a matching portion between the first deformation and the second deformation in the first repair region; A step of determining whether or not the deformation of the structure has recurred based on the determination of whether or not the matching portion exists in the first repair region; An information processing method, characterized by comprising the above.

15. A program for causing a computer to function as each means of the information processing apparatus according to any one of Claims 1 to 13.

Images