Damage information processing device, damage information processing method, program, and recording medium

Abstract

To provide a damage information processing device, a damage information processing method, and a program allowing a user to easily recognize a place having a chronologically unnatural difference.SOLUTION: A damage information processing device (10) for a structure comprises a processor (20). The processor (20) performs processing for: acquiring first damage information and second damage information at a time point chronologically later than the first damage information, the first and second damage information being damage information for a structure; extracting difference information, which is a difference between the first damage information and the second damage information; detecting a first classification location where only the first damage information is present or the first damage information is greater than the second damage information among the difference information; and outputting a notification display indicative of the first classification location on a display device in association with at least one of the first damage information and the second damage information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a damage information processing device, a damage information processing method, and a program, and in particular to a damage information processing device, a damage information processing method, and a program that process damage information of a structure. [Background technology]

[0002] Structures can suffer damage due to aging and other factors. Therefore, inspections of structures for damage are often conducted periodically. When conducting periodic inspections, it is desirable to compare past inspection results with current inspection results to understand the extent of damage, and technologies to support this have been studied.

[0003] Patent Document 1 describes a technique for acquiring two images of the same crack taken at different times and determining the change in the length and width of the crack from the two images. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-211277 Summary of the Invention [Problem to be solved by the invention]

[0005] Here, to check the degree of damage growth, the difference between two inspection results taken at different times may be extracted. When extracting this difference, various reasons may result in an unnatural difference in time series. Here, an unnatural difference in time series is, for example, when damage was present in a location in a past inspection result, but the presence of that damage cannot be confirmed in the current inspection result at the same location, even though the damage has not been repaired. Another unnatural difference in time series is, for example, when the quantitative value of the damage (length, width, or area) is smaller in the current inspection result than in the past inspection result, even though the damage has not been repaired. In other words, an unnatural difference in time series is when a result is obtained that contradicts irreversible change in time series.

[0006] Possible causes of unnatural differences in time series output like this include differences in inspection conditions (inspection workers, equipment used, environment) between past and current inspection results, improper photography of the inspection target, and improper extraction of damage from the images.

[0007] Therefore, the inspector (user) needs to recognize the locations where such unnatural differences have occurred in the time series, and take measures for the locations where the unnatural differences have occurred, such as reviewing the inspection conditions, re-photographing, re-examining the damage, etc. The above-mentioned Patent Document 1 does not mention such unnatural differences in the time series.

[0008] The present invention has been made in consideration of these circumstances, and its purpose is to provide a damage information processing device, a damage information processing method, and a program that allow a user to easily recognize areas that have unnatural differences in time series. [Means for solving the problem]

[0009] In order to achieve the above-mentioned object, one aspect of the present invention is a damage information processing device for a structure, which is equipped with a processor, and the processor acquires damage information for the structure, which is first damage information and second damage information that is chronologically later than the first damage information, extracts differential information that is the difference between the first damage information and the second damage information, detects a first classification location from the differential information where only the first damage information is present or where the first damage information is greater than the second damage information, and outputs an alarm display indicating the first classification location to a display device, associating it with at least one of the first damage information and the second damage information.

[0010] According to this aspect, differential information between the first damage information and the second damage information is extracted, and a first classification location that exists only in the first damage information is detected from the differential information. Furthermore, a first classification location where the first damage information is larger than the second damage information is detected from the differential information. Here, the first classification location is a location where only the first damage information, which is earlier in the time series, exists or a location where the first damage information is larger than the second damage information, and is a location with unnatural differential information in the time series. This aspect then outputs a notification display indicating the first classification location to the display device in association with at least one of the first damage information and the second damage information, allowing the display device to display the notification display and allowing the user to easily recognize the first classification location.

[0011] Preferably, the processor detects a second classification location from the difference information where only second damage information exists or where the second damage information is greater than the first damage information, and outputs an alarm display indicating the second classification location to the display device, associating it with at least one of the first damage information and the second damage information.

[0012] Preferably, the processor detects a third classification location where the first damage information and the second damage information overlap or where the first damage information and the second damage information are equal based on the difference information, the first damage information, and the second damage information, and outputs an alarm display indicating the third classification location to the display device in association with at least one of the first damage information and the second damage information.

[0013] Preferably, the processor switches between displaying and hiding on the display device a display related to damage information corresponding to the first classification location, damage information corresponding to the second classification location, or damage information corresponding to the third classification location.

[0014] Preferably, the processor changes the display form of the second classification location according to the magnitude of the difference included in the difference information.

[0015] Preferably, the processor performs display processing to cause the display device to display a display related to information in which the first damage information and the second damage information are superimposed.

[0016] Preferably, the processor acquires a first captured image from which the first damage information has been acquired and a second captured image from which the second damage information has been acquired, and performs display processing to display the first captured image and the second captured image side by side on a display device.

[0017] Preferably, the processor performs display processing for causing the display device to display a message recommending re-imaging when the first captured image or the second captured image does not satisfy a predetermined condition.

[0018] Preferably, the processor outputs to the display device one or more correction methods for the first damage information or the second damage information corresponding to the first classification location.

[0019] Preferably, the processor receives correction information for the first damage information or the second damage information displayed on the display device, and extracts difference information again based on the corrected first damage information or the second damage information.

[0020] Preferably, the processor automatically corrects the first damage information or the second damage information corresponding to the first classification location.

[0021] Another aspect of the present invention is a damage information processing method for a damage information processing device for a structure, which is equipped with a processor, and includes the following steps performed by the processor: acquiring damage information for the structure, which is first damage information and second damage information that is chronologically later than the first damage information; extracting differential information that is the difference between the first damage information and the second damage information; detecting a first classification location from the differential information where only the first damage information is present or where the first damage information is greater than the second damage information; and outputting an alarm display indicating the first classification location to a display device, associating it with at least one of the first damage information and the second damage information.

[0022] Another aspect of the present invention is a program that causes a structure damage information processing device having a processor to execute a damage information processing method, and causes the processor to execute the following steps: acquiring damage information for the structure, where the damage information is first damage information and second damage information that is chronologically later than the first damage information; extracting differential information that is the difference between the first damage information and the second damage information; detecting a first classification location from the differential information where only the first damage information exists or where the first damage information is greater than the second damage information; and outputting an alarm display indicating the first classification location to a display device, associating it with at least one of the first damage information and the second damage information. [Effects of the Invention]

[0023] According to the present invention, differential information between the first damage information and the second damage information is extracted, and a first classification location that exists only in the first damage information or where the first damage information is larger than the second damage information is detected from the differential information, and an alarm display indicating the first classification location is output to a display device in association with at least one of the first damage information and the second damage information, so that the display device displays the alarm display and the user can easily recognize the first classification location. [Brief explanation of the drawings]

[0024] [Figure 1]FIG. 1 is a block diagram showing an example of a hardware configuration of a damage information processing device. [Figure 2] FIG. 2 is a block diagram of the functions of the CPU. [Figure 3] FIG. 3 is a flow chart showing a display method using the damage information processing device. [Figure 4] FIG. 4 is a diagram illustrating a case where the damage information acquisition unit acquires damage information A. In FIG. [Figure 5] FIG. 5 is a diagram illustrating a case where the damage information acquisition unit acquires damage information B. In FIG. [Figure 6] FIG. 6 is a diagram illustrating alignment of captured images using feature points of the captured images. [Figure 7] FIG. 7 is a diagram illustrating extraction of differences in damage information using DP matching. [Figure 8] FIG. 8 is a diagram illustrating an example of the difference information. [Figure 9] FIG. 9 is a diagram showing an example of difference information (quantitative value). [Figure 10] FIG. 10 is a diagram showing an example of the display image (damage diagram and notification display) output by the display output unit and displayed on the display unit. [Figure 11] FIG. 11 is a diagram for explaining an example of the display form of this example. [Figure 12] FIG. 12 is a diagram illustrating the detection of cracks from a captured image by a detector. [Figure 13] FIG. 13 is a diagram illustrating an example of receiving a correction of damage information. [Figure 14] FIG. 14 is a diagram illustrating another example of receiving a correction of damage information. [Figure 15] FIG. 15 is a diagram showing the damage diagram of this example displayed on the display unit. [Figure 16] FIG. 16 is a diagram for explaining display and non-display of the damage diagram. DETAILED DESCRIPTION OF THE INVENTION

[0025] Hereinafter, preferred embodiments of an injury information processing device, an injury information processing method, and a program according to the present invention will be described with reference to the accompanying drawings.

[0026] FIG. 1 is a block diagram showing an example of a hardware configuration of the damage information processing device 10. As shown in FIG.

[0027] The damage information processing device 10 can be configured as a computer or a workstation. The hardware configuration of the damage information processing device 10 mainly includes a data acquisition unit 12, a memory 16, an operation unit 18, a CPU (Central Processing Unit) 20, a RAM (Random Access Memory) 22, and a ROM (Read Only Memory) 24. The damage information processing device 10 outputs a display image to a display unit (display device) 26. The display unit 26 is the display unit 26 of a client terminal in a client-server system connected via a network, and the damage information processing device 10 functions as a server that responds to requests from the client terminal. Therefore, the display image output from the damage information processing device 10 is displayed on the display unit 26. Alternatively, the damage information processing device 10 may be mounted on a computer, and the display unit 26 may be configured as a monitor connected to the computer.

[0028] The data acquisition unit 12 is a data input unit, and acquires information (data) stored in, for example, the memory 16. The information stored in the memory 16 may be acquired by the data acquisition unit 12, or may be stored in advance in the memory 16. For example, the data acquisition unit 12 acquires photographed images and damage information of the inspection target, which will be described later.

[0029] The memory 16 functions as a database. The memory 16 stores photographed images and damage information of the inspection target acquired by the data acquisition unit 12. For example, the memory 16 accumulates and stores photographed images and damage images that are the results of periodic inspections of structures from the past. Note that structures to be inspected include buildings, such as civil engineering structures such as bridges, tunnels, and dams, as well as buildings, houses, and architectural structures such as walls, columns, and beams of buildings. Damage to structures detected during inspections includes cracks, peeling, exposed rebar, water leakage, free lime, corrosion, and the like.

[0030] The operation unit 18 is composed of a keyboard, a pointing device such as a mouse, etc. A user inputs commands to the damage information processing device 10 via the operation unit 18.

[0031] The CPU 20 implements each function by loading a program stored in the memory 16 or the ROM 24 into the RAM 22 and executing it.

[0032] FIG. 2 is a block diagram of the functions of the CPU 20.

[0033] The CPU 20 includes a damage information acquisition unit 30, a difference extraction unit 32, a detection unit , and a display output unit .

[0034] The damage information acquisition unit 30 acquires damage information. Here, damage information is information that represents information related to damage to the structure being inspected, and is information in various forms. For example, damage information is vector information indicating damage (cracks). Also, damage information is area information of damage (peeling, exposed rebar, water leakage, free lime, corrosion).

[0035] The damage information acquisition unit 30 acquires two pieces of damage information about the same damage that were acquired at different times. The damage information acquisition unit 30 acquires, for example, damage information obtained during regular inspections that is stored in memory 16. The damage information acquisition unit 30 acquires damage information A (first damage information) acquired at time a, and damage information B (second damage information) that is damage information about the same damage as damage information A and that was acquired at time b. Note that time a is a time earlier than time b; for example, time a is August 2015 and time b is August 2020. Note that time a and time b are not particularly limited as long as they are different times.

[0036] The difference extraction unit 32 extracts difference information, which is the difference between the damage information A and the damage information B. The difference extraction unit 32 extracts the differences between the damage information A and the damage information B as difference information. For example, if the damage information A and the damage information B represent the same crack using vector information, the differences between the damage information A and the damage information B are extracted as difference information. Also, for example, if the damage information A and the damage information B represent the same peeling region using area, the differences between the damage information A and the damage information B are extracted. The difference extraction unit 32 may also extract difference information of quantitative values ​​(length, width, area) between the damage information A and the damage information B. For example, if the damage information A and the damage information B contain information regarding crack width, the difference extraction unit 32 extracts the difference information of the crack width.

[0037] The detection unit 34 detects a first classification location, which is a location in the difference information that has a difference that is unnatural in terms of time series. Here, the first classification location is a location in the difference information where only damage information A exists. Furthermore, when the damage information is expressed as a quantitative value, the first classification location is a location where damage information A is larger than damage information B. A location that exists in damage information A but not in damage information B, even though damage information A was acquired at a time point earlier in terms of time series than damage information B, is unnatural in terms of time series. Furthermore, a location where the quantitative value of damage information A is larger than the quantitative value of damage information B, even though damage information A was acquired at a time point earlier in terms of time series than damage information B, is unnatural in terms of time series. Note that damage information A and damage information B indicate that no repairs (repairs) have been performed on the detected damage.

[0038] Furthermore, the detection unit 34 detects second classification locations from the difference information, which are locations where damage has grown or appeared. Here, the second classification locations are locations in the difference information where only damage information B exists. Furthermore, when damage information is expressed as a quantitative value, the second classification locations are locations where damage information B is greater than damage information A. If damage grows or appears from time a when damage information A is acquired, only damage information B reflects the damage information corresponding to that damage.

[0039] Furthermore, based on the damage information A, the damage information B, and the difference information, the detection unit 34 detects a third classification location where the damage information A and the damage information B overlap. Furthermore, based on the damage information A, the damage information B, and the difference information, when the damage information is expressed as a quantitative value, the detection unit 34 detects a third classification location where the damage information A and the damage information B are equal. The third classification location is a damage location where no change is observed between the time when the damage information A is acquired and the time when the damage information B is acquired.

[0040] The detection unit 34 detects the second and third classification locations as necessary (such as a user setting). Therefore, the detection unit 34 can adopt a mode of detecting only the first classification location, a mode of detecting the first and second classification locations, a mode of detecting the first and third classification locations, or a mode of detecting the first, second, and third classification locations.

[0041] The display output unit 36 ​​outputs a display image to be displayed on the display unit 26. The display output unit 36 ​​outputs damage information A and damage information B and performs display processing to display the damage information A and damage information B on the display unit 26. For example, the display output unit 36 ​​displays the damage information A and damage information B on the display unit 26 in an overlapping manner. The display output unit 36 ​​also outputs a notification display indicating the first classification location to the display unit 26 in association with at least one of the displayed damage information A and damage information B, and displays the notification display on the display unit 26. In this way, by outputting the notification display indicating the first classification location to the display unit 26 and displaying the notification display on the display unit 26, the first classification location can be notified to the user, and the user can easily recognize the first classification location.

[0042] Furthermore, display output unit 36 ​​outputs a notification display indicating the second classification location to display unit 26 in association with at least one of the displayed damage information A and damage information B, and displays the notification display on display unit 26. In this way, by outputting a notification display indicating the second classification location to display unit 26 and displaying the notification display on display unit 26, the second classification location can be notified to the user, and the user can easily recognize growing or appearing damage.

[0043] Furthermore, the display output unit 36 ​​outputs a notification display indicating the third classification location to the display unit 26 in association with at least one of the displayed damage information A and damage information B, and displays the notification display on the display unit 26. In this way, by outputting a notification display indicating the third classification location to the display unit 26 and displaying the notification display on the display unit 26, the third classification location can be notified to the user, and a more accurate inspection can be performed.

[0044] Next, a description will be given of a damage information processing method using the damage information processing device 10. Fig. 3 is a flow chart showing a display method using the damage information processing device 10.

[0045] First, the damage information processing device 10 acquires damage information A and damage information B using the damage information acquisition unit 30 (step S10: damage information acquisition step). Next, the damage information processing device 10 extracts difference information between the damage information A and the damage information B using the difference extraction unit 32 (step S11: difference information extraction step). Thereafter, the damage information processing device 10 detects a first classification location from the difference information using the detection unit 34 (step S12: detection step). Note that the detection unit 34 detects a second classification location and / or a third classification location from the difference information according to user settings. Next, the damage information processing device 10 outputs a notification display indicating the first classification location to the display unit 26 using the display output unit 36 ​​(step S13: display step). Furthermore, the display output unit 36 ​​outputs a notification display indicating the second classification location and / or the third classification location to the display unit 26.

[0046] Next, each step of the above-mentioned display method will be described in detail.

[0047] <Damage information acquisition step> The damage information acquisition step is performed by damage information acquisition unit 30. Damage information acquisition unit 30 may acquire damage information, or may extract damage from an input image to generate and acquire damage information. Below, a case will be described in which a photographed image is input to damage information acquisition unit 30, and damage information acquisition unit 30 extracts damage from the photographed image to acquire damage information.

[0048] FIG. 4 is a diagram illustrating a case where the damage information acquiring unit 30 acquires the damage information A. In FIG.

[0049] FIG. 4(A) shows a photographed image (first photographed image) 50 acquired by the damage information acquisition unit 30. The damage information acquisition unit 30 acquires the photographed image 50 of the pier C photographed at time a. The damage information acquisition unit 30 then extracts cracks from the photographed image 50 using various techniques. For example, the damage information acquisition unit 30 extracts cracks from the photographed image 50 using a detector that has undergone image processing or machine learning, and acquires damage information A (vector information).

[0050] FIG. 4(B) shows damage information A acquired by the damage information acquisition unit 30. Note that the damage information A is vector information, and the damage diagram generated based on the vector information is shown. The damage information acquisition unit 30 extracts cracks captured in the captured image 50 and generates and acquires the damage information A. The damage information A is vector information and includes information on the position of the crack in the captured image 50, as well as the width and length of the crack.

[0051] FIG. 5 is a diagram illustrating a case where the damage information acquiring unit 30 acquires the damage information B. In FIG.

[0052] 5(A) shows a photographed image (second photographed image) 54 acquired by the damage information acquisition unit 30. The damage information acquisition unit 30 acquires a photographed image 54 of a pier C taken at time point b. The damage information acquisition unit 30 then extracts cracks from the photographed image 54 using various techniques. For example, the damage information acquisition unit 30 extracts cracks from the photographed image 54 using a detector that has undergone image processing or machine learning, and acquires damage information B (vector information).

[0053] FIG. 5(B) shows damage information B acquired by damage information acquisition unit 30. Note that damage information B is vector information, and the damage diagram generated based on the vector information is shown. Damage information acquisition unit 30 extracts cracks captured in captured image 54 and generates and acquires it as damage information B. Damage information B is vector information and includes information on the position of the crack in captured image 54, as well as information on the width and length of the crack.

[0054] If the input photographed images 50 and 54 are not suitable for obtaining damage information (if they do not satisfy predetermined conditions), the display output unit 36 ​​may output a message recommending re-imaging to the display unit 26, and cause the display unit 26 to display a message recommending re-imaging. The user can see this message and re-input the re-imaging image.

[0055] As described above, in the damage information acquisition step, the damage information acquisition unit 30 acquires the damage information A and the damage information B.

[0056] <Difference information extraction step> The difference information extraction step is performed by the difference extraction unit 32. The difference extraction unit 32 uses various methods to extract the difference between the damage information A and the damage information B. An example of the method for extracting difference information by the difference extraction unit 32 will be described below.

[0057] (Example 1) In Example 1, difference extraction unit 32 uses image features to align damage information A and damage information B and extracts difference information between damage information A and damage information B. In the example described above, damage information acquisition unit 30 acquires photographed image 50 to acquire damage information A, and photographed image 54 to acquire damage information B. In such a case, difference extraction unit 32 aligns photographed image 50 and photographed image 54, and then aligns damage information A and damage information B based on the aligned information.

[0058] FIG. 6 is a diagram illustrating alignment of the photographed image 50 and the photographed image 54 using feature points of the photographed image 50 and the photographed image 54. In FIG.

[0059] First, the difference extraction unit 32 extracts feature points from the captured image 50 and the captured image 54. For example, the difference extraction unit 32 extracts feature points from the captured image 50 and the captured image 54 using feature point detection techniques such as ORB (Oriented Fast and Rotated Brief) or AKAZE (Accelerated-KAZE). FIG. 6 shows one feature point PA among the multiple feature points extracted from the captured image 50. Also, one feature point PB among the multiple feature points extracted from the captured image 54 is shown.

[0060] Thereafter, the difference extraction unit 32 derives corresponding feature points between the photographed image 50 and the photographed image 54. The difference extraction unit 32 derives corresponding points between the photographed image 50 and the photographed image 54 using, for example, Brue-Force or FLANN (Fast Library for Approximate Nearest Neighbors) technology. In the case shown in Fig. 6, the correspondence between the feature points PA and PB is indicated by a line COR.

[0061] As described above, difference extraction unit 32 derives the correspondence between multiple feature points in photographed image 50 and multiple feature points in photographed image 54 (detection of corresponding points). Then, difference extraction unit 32 aligns photographed image 50 with photographed image 54 using the detected corresponding points. Specifically, from the correspondence between photographed image 50 and photographed image 54, a coordinate transformation model (coordinate transformation matrix) between coordinates on photographed image 50 and coordinates on photographed image 54 is generated. Then, difference extraction unit 32 aligns damage information A with damage information B using the coordinate transformation model, and extracts the difference between the aligned damage information A and damage information B.

[0062] (Example 2) In Example 2, the difference extraction unit 32 aligns the damage information A and the damage information B using a method that uses dynamic programming (DP (Dynamic Programming) matching), and extracts the difference information between the damage information A and the damage information B. Note that the DP matching alignment method is preferably used to extract the difference in damage information related to cracks.

[0063] FIG. 7 is a diagram illustrating extraction of differences in damage information using DP matching.

[0064] The damage information AP and the damage information BP are damage information for the same crack at different times. The damage information AP and the damage information BP are each vector information indicating the crack. The difference extraction unit 32 uses DP matching to derive corresponding locations a to d between the damage information AP and the damage information BP. Then, the difference extraction unit 32 extracts difference information between the damage information AP and the damage information BP at each of the corresponding locations a to d.

[0065] (Example 3) In Example 3, difference extraction unit 32 aligns damage information A and damage information B using a method that uses AI (Artificial Intelligence), and extracts difference information between damage information A and damage information B. For example, difference extraction unit 32 aligns photographed image 50 and photographed image 54 using a recognizer that aligns images and that has been trained (generated) using deep learning, and then aligns damage information A and damage information B based on the positional information after the alignment, and extracts difference information.

[0066] As described above, the difference extracting unit 32 aligns the damage information A and the damage information B using various methods, and extracts the difference between the damage information A and the damage information B after the alignment.

[0067] <Detection step> The detection step is performed by the detection unit 34. The detection unit 34 detects first classified locations from the difference information, which are locations having only damage information A or locations where damage information A is greater than damage information B. The detection unit 34 also detects second classified locations from the difference information, which are locations having only damage information B or locations where damage information B is greater than damage information A. The detection unit 34 also detects third classified locations, which are locations having damage information A and damage information B or locations where damage information A and damage information B are equal, based on the difference information, damage information A, and damage information B.

[0068] Furthermore, the detection unit 34 detects, as a first classification location, a location where the damage information A is greater than the damage information B in the quantitative value of the damage information. The detection unit 34 also detects, as a second classification location, a location where the damage information B is greater than the damage information A in the quantitative value of the damage information. The detection unit 34 also detects, as a third classification location, a location where the damage information A and the damage information B are equal in the quantitative value of the damage information.

[0069] FIG. 8 is a diagram showing an example of difference information between damage information A and damage information B. As shown in FIG.

[0070] Difference information 52 is obtained by aligning damage information A and damage information B and subtracting damage information A from damage information B. The illustrated difference information 52 is displayed as vector information obtained by subtracting damage information A from damage information B, damage information A, and a damage diagram generated based on the damage information B. A portion Q (surrounded by a dotted line) in difference information 52 is a portion where only damage information A exists or where damage information A is greater than damage information B if the damage information is expressed as a quantitative value, and is detected by detection unit 34 as a first classification portion. A portion R (surrounded by a solid black line) in difference information 52 is a portion where only damage information B exists or where damage information B is greater than damage information A if the damage information is expressed as a quantitative value, and is detected by detection unit 34 as a second classification portion. Furthermore, location S (dotted area) in the differential information 52 is a location where damage information A and damage information B exist, or where damage information A and damage information B are equal if the damage information is expressed as a quantitative value, and the detection unit 34 detects this as a third classification location.

[0071] In this way, the detection unit 34 detects the first classification location and the second classification location from the difference information 52. Furthermore, the detection unit 34 detects the third classification location based on the damage information A, the damage information B, and the difference information 52.

[0072] FIG. 9 is a diagram showing an example of difference information (quantitative value) between damage information A and damage information B. As shown in FIG.

[0073] Fig. 9 explains classification based on quantitative values ​​related to crack widths (when damage information is expressed as quantitative values) in the difference information 52 shown in Fig. 8. Note that the difference information shown in the figure shows a numerical value obtained by subtracting the crack width of damage information A from the crack width of damage information B.

[0074] Enlarged view 66 and enlarged view 68 are enlarged views of a portion of the difference information 52. For the crack (number D96_3), the difference information is -0.15, and the detection unit 34 detects the crack (number D96_3) as a first classification location. For the crack (number D97_3), the difference information is 0.05, and the detection unit 34 detects the crack (number D97_3) as a second classification location. For the crack (number D14_1), the difference information is 0.00, and the detection unit 34 detects the crack (number D14_1) as a third classification location.

[0075] In this way, the detection unit 34 detects the first classification location and the second classification location from the difference information regarding the crack width. Also, the detection unit 34 detects the third classification location from the damage information A, the damage information B, and the difference information 52.

[0076] As described above, the detection unit 34 detects the first classification location, the second classification location, and the third classification location according to the locations of the damage information A and the damage information B in the difference information 52. Furthermore, when the difference information 52 has a quantitative value, the detection unit 34 detects the first classification location, the second classification location, and the third classification location according to the magnitude relationship (including the case where the damage information A and the damage information B are equal) between the damage information A and the damage information B.

[0077] <Display steps> The display step is performed by the display output unit 36. The display output unit 36 ​​outputs a notification display indicating the first classification location to the display unit 26, and performs display processing to cause the display unit 26 to display the notification display. The display output unit 36 ​​also outputs a notification display indicating the second classification location to the display unit 26, and performs display processing to cause the display unit 26 to display the notification display. The display output unit 36 ​​also performs display processing to cause the display unit 26 to display a notification display indicating the third classification location.

[0078] FIG. 10 is a diagram showing an example of the display image (damage diagram and notification display) output by the display output unit 36 ​​displayed on the display unit 26. In FIG.

[0079] Display unit 26 displays a damage diagram 53 generated based on damage information A and damage information B. In damage diagram 53, the first classification location is displayed with a white solid line notification indicator M. In damage diagram 53, the second classification location is displayed with a black solid line notification indicator N. In damage diagram 53, the third classification location is displayed with a dotted line notification indicator P. In this way, in damage diagram 53 displayed on display unit 26, notification indicators are displayed for the first classification location, the second classification location, and the third classification location using different colors and line types.

[0080] Furthermore, display unit 26 displays a white balloon-shaped notification display V1 at the first classification location. Notification display V1 is displayed at a position that notifies the first classification location based on damage information A detected as the first classification location. Display unit 26 also displays a black balloon-shaped notification display V2 at the first classification location where damage information A has a larger quantitative value than damage information B. In this way, display unit 26 displays a notification display for the first classification location using a marker such as a balloon on damage diagram 53.

[0081] As described above, according to this embodiment, difference information 52 between damage information A and damage information B is extracted, and a first classification location is detected from the difference information 52. Then, this embodiment displays a notification indicating the first classification location on display unit 26 in association with at least one of damage information A and damage information B, allowing the user to recognize locations having difference information that is unnatural in chronological order.

[0082] In this embodiment, the second classification location is detected from the difference information 52. Then, in this embodiment, a notification display indicating the second classification location in association with at least one of the damage information A and the damage information B is displayed on the display unit 26, thereby allowing the user to recognize the location where the damage has grown or appeared.

[0083] In this embodiment, the third classification location is detected based on the difference information 52, the damage information A, and the damage information B. Then, in this embodiment, a notification indicating the third classification location is displayed on the display unit 26 in association with at least one of the damage information A and the damage information B, thereby allowing the user to recognize the location where the damage has not changed.

[0084] <First Modification> Next, a first modification of the above-described embodiment will be described. In this modification, a captured image 62 related to damage information A and a captured image 64 related to damage information B, which correspond to an area designated by the user, are displayed on the display unit 26.

[0085] 11A and 11B are diagrams illustrating an example of the display form of this example. Fig. 11A is a diagram showing a damage diagram 53 displayed on the display unit 26.

[0086] The user designates a partial area F in the damage diagram 53 displayed on the display unit 26. The display output unit 36 ​​accepts the designation of the partial area and outputs the photographed image 62 and the photographed image 64 corresponding to the designated area to the display unit 26.

[0087] FIG. 11(B) is a diagram showing the photographed image 62 and the photographed image 64 displayed on the display unit 26.

[0088] The photographed image 62 is an enlarged view of an area corresponding to area F in the photographed image 50. The photographed image 64 is an enlarged view of an area corresponding to area F in the photographed image 54. The photographed image 62 and the photographed image 64 are output from the display output unit 36 ​​to the display unit 26. Then, the display output unit 36 ​​performs a display process to display the photographed image 62 and the photographed image 64 side by side so that they can be compared (comparison display) on the display unit 26. Note that the photographed image 62 and the photographed image 64 may be displayed with corresponding damage diagrams superimposed thereon.

[0089] In this way, by displaying on the display unit 26 a past photographed image 62 and a current photographed image 64 of the area specified by the user, the user can compare and observe the state of the inspection target in the past with the state of the inspection target in the current state.

[0090] <Second Modification> Next, a second modification of the above embodiment will be described. In this modification, a method for correcting the first classified portion is presented to the user.

[0091] Damage information corresponding to the first classification location (damage information A or damage information B) can be corrected by various methods. Therefore, the display output unit 36 ​​outputs one or more correction methods to the display unit 26. The display unit 26 then displays the input correction method and presents the correction method to the user. Specific examples of the correction methods will be described below.

[0092] Damage information corresponding to the first classification location can be corrected using the inspection date and time and the temperature at the time of inspection. For example, since the volume of concrete changes with temperature, the temperature is estimated from the inspection date and time and image features, and the quantitative value (crack width) in the damage information is corrected. In addition, the user can input the date and time and temperature to correct the quantitative value in the damage information.

[0093] Furthermore, when the damage information of the first classification location is generated by a detector that has undergone machine learning, the damage information can be corrected by changing the threshold value or the like.

[0094] FIG. 12 is a diagram illustrating the detection of cracks from a captured image by a detector.

[0095] 12(A) shows a captured image 70 of an inspection target. The captured image 70 is input to a detector (AI) that has been trained by machine learning, which detects cracks in the captured image 70 and outputs a detection result 72. The detector (AI) outputs the probability that it is a crack, and detects those with a probability above a predetermined threshold as damage.

[0096] FIG. 12(B) shows the detection result 72 output from the detector (AI). For example, the detector (AI) generates the detection result 72 by determining that an area with a crack probability of 30% or more is a crack. In the detection result 72, areas with a high crack probability are represented by a dark line (indicated by arrow I), and areas with a low crack probability are represented by a light line (indicated by arrow H). In the detection result 72, the crack probability is displayed as a pixel value.

[0097] When correcting a first classified location where only damage information A exists or where damage information A is greater than damage information B, the damage detection threshold when generating damage information B is lowered and the detector (AI) is made to redetect damage. As a result, damage corresponding to damage information A is also detected in damage information B, and damage information A and damage information B are present in the location detected as the first classified location, or the quantitative value of damage information B is corrected, and the first classified location is corrected. Alternatively, when correcting a first classified location where only damage information A exists or where damage information A is greater than damage information B, the damage detection threshold when generating damage information A is raised and the detector (AI) is made to redetect damage. As a result, damage information A and damage information B are no longer present in the location detected as the first classified location, or the quantitative value of damage information A is corrected, and the first classified location is corrected.

[0098] As described above, the display output unit 36 ​​outputs one or more correction methods for the damage information (damage information A or damage information B) corresponding to the first classification location to the display unit 26, and the display unit 26 displays the correction methods on the display unit 26, thereby presenting the correction methods to the user.

[0099] <Third Modification> Next, a third modification of the above-described embodiment will be described. In this example, the user manually corrects the damage information A or the damage information B, and the difference extraction process is performed again. As described above, the damage information corresponding to the first classification location (damage information A or damage information B) is corrected. In this example, this correction is performed manually by the user.

[0100] FIG. 13 is a diagram illustrating an example of receiving a correction of damage information.

[0101] When a damage diagram 80 of a crack is displayed on the display unit 26 based on the damage information, the user adds a damage diagram 82 of a crack using a pointing device. The addition of the damage diagram 82 is accepted as correction information by the damage information acquisition unit 30. Then, the damage information acquisition unit 30 corrects the existing damage information based on the correction information, and then the difference extraction unit 32 extracts the difference information again.

[0102] FIG. 14 is a diagram illustrating another example of receiving a correction of damage information.

[0103] When a damage diagram 84 of a damaged area (e.g., peeling) is displayed on display unit 26 based on the damage information, the user adds a damage diagram 86 of the damaged area using the pointing device. The addition of damage diagram 86 is accepted as correction information by damage information acquisition unit 30. Then, damage information acquisition unit 30 corrects the existing damage information based on the correction information, and then difference extraction unit 32 extracts difference information again.

[0104] In this way, in the case of linear damage such as cracks, the user can correct the damage information by adding or deleting trace lines, etc. In the case of planar damage such as peeling, the user can correct the damage information by adding or deleting areas, etc.

[0105] In the above example, the damage information is manually corrected by the user, but the damage information may be automatically corrected. For example, the damage information acquisition unit 30 may automatically correct the damage information by deleting damage information A from the location detected as the first classification location or by adding damage as damage information B to the location detected as the first classification location.

[0106] <Fourth Modification> Next, a fourth modification of the above embodiment will be described. In this modification, the display of the damage diagram is changed depending on the magnitude of the difference information (quantitative value).

[0107] 15 is a diagram showing the damage diagram of this example displayed on the display unit 26. The display unit 26 is controlled by the display output unit 36 ​​so as to display the display image output from the display output unit 36.

[0108] Damage diagram 55 is a damage diagram generated based on damage information corresponding to the second classification location. Damage diagram 55 changes the line type of the damage diagram showing the crack depending on the magnitude of the difference in the quantitative value difference information (crack width).

[0109] The damage diagram indicated by arrow W corresponds to differential information with a large amount of crack width growth. This damage diagram is displayed using a thin dotted line. The damage diagram indicated by arrow X corresponds to differential information with a medium amount of crack width growth. This damage diagram is displayed using a thick dotted line. The damage diagram indicated by arrow Y corresponds to differential information with a small amount of crack width growth. This damage information is displayed using a solid line.

[0110] In this way, by changing the display format depending on the magnitude of the quantitative value in the difference information, it is possible to notify the user of dangerous locations where damage is growing.

[0111] <Fifth Modification> Next, a fifth modification of the above-described embodiment will be described. In this modification, control is performed to switch between displaying and hiding damage diagrams corresponding to the first classification location, the second classification location, and the third classification location.

[0112] 16 is a diagram for explaining display and non-display of a damage diagram corresponding to the first classification location on the display unit 26. The display unit 26 is controlled by the display output unit 36 ​​so as to display the display image output from the display output unit 36.

[0113] 16(A) is a diagram showing a damage diagram 53 displayed on the display unit 26. The damage diagram 53 displays all of the damage diagrams generated from the damage information corresponding to the first classification location, the second classification location, and the third classification location. Note that the damage diagram corresponding to the first classification location is displayed with a white line (indicated by arrow Z).

[0114] 16(B) is a diagram showing a damage diagram 53 in which the damage diagram (arrow Z) corresponding to the first classification location is hidden. The display output unit 36 ​​outputs the display image in which the first classification location is hidden to the display unit 26, and performs display processing to cause the display unit 26 to display the image.

[0115] In this way, by controlling the display or non-display of the damage diagram corresponding to the first classification location, the second classification location, or the third classification location on the display unit 26, the user can be provided with the desired display of the damage diagram.

[0116] <Other> In the above embodiment, the hardware structure of the processing unit that executes various processes is the following various processors: The various processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) and functions as various processing units, a programmable logic device (PLD), such as an FPGA (Field Programmable Gate Array), whose circuit configuration can be changed after manufacture, and a dedicated electrical circuit, such as an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing specific processes.

[0117] A single processing unit may be configured with one of these various processors, or may be configured with two or more processors of the same or different types (for example, multiple FPGAs, or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with a single processor. Examples of multiple processing units configured with a single processor include, first, a configuration in which one processor is configured with a combination of one or more CPUs and software, as typified by computers such as client and server, and this processor functions as multiple processing units. Second, a configuration in which a processor is used to realize the functions of an entire system including multiple processing units on a single IC (Integrated Circuit) chip, as typified by a system-on-chip (SoC). In this way, the various processing units are configured with one or more of the above-mentioned various processors as a hardware structure.

[0118] Furthermore, the hardware structure of these various processors is, more specifically, an electric circuit made up of a combination of circuit elements such as semiconductor elements.

[0119] The above-described configurations and functions can be realized by any hardware, software, or a combination of both. For example, the present invention can be applied to a program that causes a computer to execute the above-described processing steps (processing procedures), a computer-readable recording medium (non-transitory recording medium) on which such a program is recorded, or a computer on which such a program can be installed.

[0120] Although examples of the present invention have been described above, it goes without saying that the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the present invention. [Explanation of symbols]

[0121] 10: Damage information processing device 12: Data acquisition section 16: Memory 18:Operation section 20: CPU 22:RAM 24:ROM 26:Display section 30: Damage information acquisition department 32: Difference extraction part 34: Detection unit 36: Display output section

Claims

1. A structure damage information processing device including a processor, The processor: acquiring first damage information and second damage information at a time later in time series than the first damage information, the second damage information being vector information or damage area information indicating damage to the structure; extracting difference information that is a difference between the first damage information and the second damage information; Detecting a portion of the difference information where only the second damage information exists as a second classification portion; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; Damage information processing device.

2. A structure damage information processing device including a processor, The processor: Damage information is vector information indicating cracks in a structure, the damage information having information on the width of the cracks, and includes first damage information and second damage information at a time later in time than the first damage information; Extracting difference information that is a location where the width of the crack differs between the first damage information and the second damage information; Detecting, from the difference information, a portion where the width of the crack in the second damage information is larger than the width of the crack in the first damage information as a second classification portion; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; Damage information processing device.

3. The processor: detecting a third classification location where the first damage information and the second damage information overlap or where the first damage information and the second damage information are equal based on the difference information, the first damage information, and the second damage information; The damage information processing device according to claim 1 or 2, wherein a display image in which a notification display indicating the third classification location is performed in the damage diagram is output to the display device.

4. The damage information processing device according to claim 3 , wherein the processor switches between displaying and hiding the damage diagram corresponding to the second classification location or the damage diagram corresponding to the third classification location on the display device in the display image.

5. The damage information processing device according to claim 2 , wherein the processor changes the display form of the notification display indicating the second classification location depending on the magnitude of the difference in width of the crack.

6. the processor acquires a first photographed image from which the first damage information has been acquired and a second photographed image from which the second damage information has been acquired; The damage information processing device according to claim 1 , further comprising a display process for displaying the first photographed image and the second photographed image side by side on the display device.

7. The damage information processing device according to claim 6, wherein the processor performs display processing to cause the display device to display a message recommending re-imaging if the first captured image or the second captured image does not satisfy a predetermined condition.

8. The processor: The damage information processing device according to claim 1 , further comprising: outputting, to the display device, one or more correction methods for the first damage information or the second damage information corresponding to the second classification location.

9. The processor: receiving correction information for the first damage information or the second damage information displayed on the display device; The damage information processing device according to claim 1 , wherein the difference information is extracted again based on the corrected first damage information or the corrected second damage information.

10. The damage information processing device according to claim 1 , wherein the processor automatically corrects the first damage information or the second damage information corresponding to the second classification location.

11. A damage information processing method for a structure damage information processing device including a processor, The processor performs: acquiring first damage information and second damage information at a time later in time series than the first damage information, the damage information being vector information or damage area information indicating damage to the structure; extracting difference information that is a difference between the first damage information and the second damage information; detecting a portion in the difference information where only the second damage information exists as a second classification portion; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; A damage information processing method including:

12. A damage information processing method for a structure damage information processing device including a processor, The processor performs: acquiring first damage information and second damage information at a time later in time series than the first damage information, the second damage information being vector information indicating cracks in a structure and having information on the width of the cracks; Extracting difference information that is a location where the width of the crack differs between the first damage information and the second damage information; detecting, from the difference information, a location where the width of the crack in the second damage information is larger than the width of the crack in the first damage information as a second classification location; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; A damage information processing method including:

13. A program for causing a structure damage information processing device to execute a damage information processing method, the program comprising: the processor, acquiring first damage information and second damage information at a time later in time series than the first damage information, the damage information being vector information or damage area information indicating damage to the structure; extracting difference information that is a difference between the first damage information and the second damage information; detecting a portion in the difference information where only the second damage information exists as a second classification portion; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; A program that executes the following.

14. A program for causing a structure damage information processing device to execute a damage information processing method, the program comprising: the processor, acquiring first damage information and second damage information at a time later in time series than the first damage information, the second damage information being vector information indicating cracks in a structure and having information on the width of the cracks; Extracting difference information that is a location where the width of the crack differs between the first damage information and the second damage information; detecting, from the difference information, a location where the width of the crack in the second damage information is larger than the width of the crack in the first damage information as a second classification location; outputting, to a display device, a display image in which a notification display indicating the second classification location is performed in a damage diagram generated based on the first damage information and the second damage information; A program that executes the following.

15. A non-transitory computer-readable recording medium on which the program according to claim 13 or 14 is recorded.

Images