Structural inspection support system, structural inspection support device, structural inspection support method, and program

Abstract

To improve the efficiency of structural inspections. [Solution] A structural inspection support system is provided, which includes: an acquisition means for acquiring 3D data of a structure to be inspected and composed of multiple members, and an inspection damage imaging image taken of damage to the structure during inspection; and a determination means for classifying the 3D data for each of the multiple members, determining the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the degree of damage based on the 3D data and the inspection damage imaging image, and determining the soundness of the damage by inputting the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the degree of damage into a diagnostic model.

Description

Technical Field

[0001] The present disclosure relates to a structure inspection support system, a structure inspection support device, a structure inspection support method, and a program.

Background Art

[0002] Bridge maintenance is generally carried out in three steps: an inspection step, a diagnosis step, and a repair step. In the inspection step, it is necessary to determine the location, member, type, and degree of damage on the bridge. In particular, in order to determine the degree of damage, it is necessary to take pictures of the damage, visually check it, and measure the damage dimensions. Especially, in order to measure the dimensions of damage occurring at high places, it is necessary to use a ladder or an aerial work platform and measure using a crack gauge, which requires time and labor. Also, when a huge number of damages are recognized for one bridge, even only the inspection step will require a huge amount of time.

[0003] In the inspection step, further, the creation of an inspection report and a two-dimensional damage diagram defined by law is required. The inspection report records the location, member, type, degree, and dimensions of the damage. These inspection reports and two-dimensional damage diagrams are mainly referred to when repairing the bridge and at the time of the next inspection. The efficient creation of these inspection reports and two-dimensional damage diagrams has become an urgent issue.

[0004] As this type of technology, Patent Documents 1 to 3 are known.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

[0006] The purpose of this disclosure is to improve the efficiency of structural inspections. [Means for solving the problem]

[0007] We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, A structural inspection support system will be provided.

[0008] We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, A support device for structural inspection will be provided.

[0009] A computer acquires three-dimensional data of a structure to be inspected and composed of a plurality of members, acquires an inspection-time damage imaging image that images the damage of the structure during the inspection of the structure, classifies the three-dimensional data for each of the plurality of members, based on the three-dimensional data and the inspection-time damage imaging image, determines the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the dimension of the damage, and the degree of the damage, and determines the soundness of the damage by inputting the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the dimension of the damage, and the degree of the damage into a diagnostic model. A method for supporting the inspection of a structure is provided.

[0010] For a computer, acquires three-dimensional data of a structure to be inspected and composed of a plurality of members, acquires an inspection-time damage imaging image that images the damage of the structure during the inspection of the structure, an acquisition means, classifies the three-dimensional data for each of the plurality of members, based on the three-dimensional data and the inspection-time damage imaging image, determines the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the dimension of the damage, and the degree of the damage, and determines the soundness of the damage by inputting the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the dimension of the damage, and the degree of the damage into a diagnostic model. a determination means, A program for causing the computer to function as such is provided.

Advantages of the Invention

[0011] According to the present disclosure, it is possible to improve the inspection efficiency of a structure.

Brief Description of the Drawings

[0012] [Figure 1]It is a block diagram of a structure inspection support system. [Figure 2] It is an operation flow of a structure inspection support system. [Figure 3] It is a block diagram of a structure inspection support device. [Figure 4] It is an operation flow of a structure inspection support device. [Figure 5] It is an operation flow of a structure inspection support device. [Figure 6] It is an operation flow of a structure inspection support device. [Figure 7] It is an operation flow of a structure inspection support device. [Figure 8] It is a diagram showing the criteria for judging the degree of floor slab cracks.

Embodiments for Carrying Out the Invention

[0013] (Summary of the Present Disclosure) First, the summary of the present disclosure will be described. FIG. 1 shows a block diagram of a structure inspection support system 1000. As shown in FIG. 1, the structure inspection support system 1000 includes an acquisition means 1001 and a determination means 1002.

[0014] The acquisition means 1001 acquires three-dimensional data of a structure that is an inspection target and is composed of a plurality of members. The acquisition means 1001 acquires an inspection-time damage imaging image obtained by imaging the damage of the structure during the inspection of the structure.

[0015] The determination means 1002 classifies the three-dimensional data for each of the plurality of members. The determination means 1002 determines the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the size of the damage, and the degree of the damage based on the three-dimensional data and the inspection-time damage imaging image. The determination means 1002 determines the soundness of the damage by inputting the three-dimensional position of the damage, the type of the member to which the damage belongs, the type of the damage, the size of the damage, and the degree of the damage into a diagnostic model.

[0016] Next, the operation of the structure inspection support system 1000 will be described. FIG. 2 is an operation flow of the structure inspection support system 1000.

[0017] First, the acquisition means 1001 acquires three-dimensional data of the structure to be inspected, which is composed of multiple members (S1001). Next, the acquisition means 1001 acquires inspection damage images, which are images of damage to the structure taken during the inspection of the structure (S1002).

[0018] Next, the determination means 1002 classifies the 3D data for each of the multiple members (S1003). Next, the determination means 1002 determines the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage based on the 3D data and the damage image taken during inspection (S1004). Next, the determination means 1002 determines the soundness of the damage by inputting the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into the diagnostic model (S1005).

[0019] According to this disclosure, it is possible to improve the efficiency of structural inspections.

[0020] (Embodiment) Next, a first embodiment of the present disclosure will be described. Figure 3 is a block diagram of a structural inspection support device 1. The structural inspection support device 1 is a specific example of a structural inspection support system. As shown in Figure 3, the structural inspection support device 1 includes a processor 1a, a memory 1b, a communication interface 1c, and an LCD 1d (Liquid Crystal Display). The structural inspection support device 1 may be composed of a single device or may be realized by distributed processing by multiple devices. The processor 1a has access to the memory 1b. The processor 1a communicates with other devices via the communication interface 1c. The processor 1a reads and executes a program stored in the memory 1b. In this way, the processor 1a causes the hardware, such as the processor 1a, to function as an acquisition unit 2, a determination unit 3, a generation unit 4, and an output unit 5. The memory 1b stores various types of data used for information processing by the processor 1a. Specifically, the memory 1b stores a point cloud DB 6 and an image DB 7.

[0021] The structural inspection support device 1 is a device for speeding up and streamlining the inspection of structures. In this embodiment, a bridge is used as an example of a structure to be inspected. However, the structure may be a substation, radio tower, dam, or other structure. A bridge is composed of multiple members. These members typically include a deck slab, main girders, bearings, and the main frame.

[0022] The acquisition unit 2 acquires 3D data of the bridge. Specifically, the acquisition unit 2 acquires 3D data of the bridge and its surrounding facilities. Surrounding facilities typically include roadways, sidewalks, and other facilities that exist around the bridge. Hereinafter, the bridge and its surrounding facilities may be simply referred to as the bridge. The 3D data is typically a point cloud. That is, the acquisition unit 2 reads and acquires the point cloud from a recording medium storing the point cloud each time an inspection is performed. Alternatively, the acquisition unit 2 may acquire the point cloud via the communication interface 1c from a point cloud generation device that generates a point cloud by measuring the distance of the bridge and its surrounding facilities.

[0023] To ensure a comprehensive point cloud of the bridge, the point cloud generator measures the distance of the bridge from multiple locations. The acquisition unit 2 then synthesizes the multiple sets of point clouds acquired from the point cloud generator, typically using ICP (Iterative Closest Point).

[0024] Point cloud generation devices are typically LiDAR (Light Detection and Ranging) devices. However, they can also be replaced by Radar (Radio Detection and Ranging) devices or stereo cameras.

[0025] The acquisition unit 2 stores the acquired point cloud, or the combined point cloud, in the point cloud DB6.

[0026] Furthermore, the acquisition unit 2 acquires inspection damage images taken when bridge damage is detected during bridge inspections. Typically, bridge inspectors use an imaging device to image damage they discover visually. The acquisition unit 2 then acquires the inspection damage images from the imaging device via the communication interface 1c. The acquisition unit 2 registers the acquired inspection damage images in the image database 7. The image database 7 also contains past damage images, which are inspection damage images from past inspections. Hereinafter, damage included in inspection damage images may be referred to as inspection damage, and damage included in past damage images may be referred to as past damage.

[0027] The determination unit 3 classifies the bridge point cloud into multiple members. Typically, the determination unit 3 uses semantic segmentation to color-code the bridge point cloud with different colors for each member and also labels each member.

[0028] The determination unit 3 determines the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage, based on the bridge point cloud and damage images taken during inspection. The determination unit 3 then inputs the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage into a diagnostic model to determine the soundness of the damage.

[0029] The diagnostic model is a pre-trained model that takes the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage as input, and outputs the soundness of the damage.

[0030] The three-dimensional location of damage is typically the three-dimensional coordinates in the point cloud coordinate system of the damage.

[0031] The type of member to which the damage belongs is typically the same as the type of member in which the damage is present.

[0032] The types of damage typically include cracks, spalling, exposed rebar, and other deformations.

[0033] The dimensions of the damage are typically the dimensions of the damage in the point cloud coordinate system.

[0034] The degree of damage refers, for example, to a five-point scale. Typically, the degree of damage is assessed based on the type and dimensions of the damage.

[0035] The progression of damage indicates whether the size of the damage has increased compared to the past.

[0036] The criteria for determining damage include additional information about the damage, such as whether or not there is a concern about rebar corrosion.

[0037] The soundness of damage refers to a four-level evaluation of the damage's condition. Simply put, the soundness of damage indicates the urgency of repairing the damage. For example, when the soundness is Level I (sound), it means that the bridge's function is not impaired. When the soundness is Level II (preventive maintenance stage), it means that the bridge's function is not impaired, but it is desirable to take measures from a preventive maintenance perspective. When the soundness is Level III (early action stage), it means that there is a possibility of the bridge's function being impaired, and measures should be taken as soon as possible. When the soundness is Level IV (emergency action stage), it means that the bridge's function is impaired, or there is a very high possibility of it being impaired, and measures should be taken urgently.

[0038] The generation unit 4 generates an inspection report and a two-dimensional damage diagram using the inspection results, which include the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, as well as a point cloud.

[0039] The generation unit 4 generates inspection reports in accordance with the inspection report format established by the Ministry of Land, Infrastructure, Transport and Tourism. Typically, an inspection report is a document that summarizes the images of the damage taken during the inspection and the inspection results for each type of damage.

[0040] The generation unit 4 generates a two-dimensional damage diagram in accordance with the two-dimensional damage diagram format specified by the Ministry of Land, Infrastructure, Transport and Tourism. A two-dimensional damage diagram is typically a drawing of a bridge cross-section with inspection results indicated for each type of damage.

[0041] Inspection reports and 2D damage diagrams are referenced when repairing damage, when planning damage repairs, and during the next inspection.

[0042] Furthermore, the generation unit 4 generates 3D data for viewing results by superimposing inspection results, including the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, onto the 3D data. Typically, the generation unit 4 generates 3D data for viewing results by superimposing inspection results for each type of damage onto the point cloud of the bridge. At this time, the generation unit 4 may generate the 3D data for viewing results in such a way that the areas where damage exists are colored and highlighted.

[0043] Output unit 5 is a specific example of an inspection candidate position output means. Output unit 5 calculates the 3D position of past damage based on the bridge point cloud and past damage image data, and outputs the 3D position of past damage superimposed on the point cloud as an inspection candidate position to LCD 1d. By viewing the point cloud, inspectors can easily grasp the inspection candidate positions during inspection.

[0044] The output unit 5 displays the 3D data for viewing the results on the LCD 1d. This allows the point cloud of the bridge, with the inspection results superimposed, to be viewed from various angles after the inspection.

[0045] The output unit 5 outputs the inspection report and 2D damage diagram generated by the generation unit 4 to a recording medium or the like.

[0046] Next, the operation of the structural inspection support device 1 will be explained in detail with reference to Figures 4 to 7.

[0047] <Preparation before inspection> First, as shown in Figure 4, the acquisition unit 2 acquires multiple sets of point clouds of the bridge as preliminary preparation before inspection (S100). Next, the acquisition unit 2 combines the multiple sets of point clouds by aligning them (S110). Next, the judgment unit 3 classifies the bridge point clouds into multiple members (S120). At this time, the judgment unit 3 may also color-code the bridge point clouds according to the multiple members.

[0048] <During inspection> Next, as shown in Figure 5, the output unit 5 acquires past damage images stored in the image DB7 during inspection (S200). Next, the output unit 5 determines the three-dimensional position and dimensions of the past damage and saves the determination result to the memory 1b (S210). Specifically, the determination unit 3 determines the three-dimensional position and dimensions of the past damage by comparing the past damage images with the point cloud of the bridge.

[0049] Furthermore, the comparison between past damage images and the bridge point cloud can be performed by various methods. For example, when an inspector images the damage, they take images of the damage from both a close-up and a long-distance perspective. The determination unit 3 then determines the position and range of the close-up image in the long-distance image, and compares the long-distance image with the bridge point cloud. Using this comparison result, the determination unit 3 can perform a comparison between the close-up image of the damage and the bridge point cloud.

[0050] Identifying areas of past damage in previously acquired damage images can be done by various methods. For example, a damage analysis model, which is a pre-trained model that takes previously acquired damage images as input and outputs the areas of damage contained in those images, can be used. Identifying areas of damage at the time of inspection in previously acquired damage images can be done by similar methods.

[0051] Next, the output unit 5 acquires the 3D location of past damage from memory 1b (S300). Then, the output unit 5 overlays the 3D location of past damage as a candidate inspection location onto the bridge point cloud and displays it on LCD 1d (S310). By viewing this point cloud, the inspector can easily identify the candidate inspection locations for the current inspection. The inspector identifies the candidate inspection locations by viewing the point cloud and images the damage in question during the current inspection. In addition, the inspector images any damage newly discovered during the current inspection.

[0052] Next, as shown in Figure 7, the acquisition unit 2 acquires images of bridge damage during inspection (S400). Next, the determination unit 3 inputs the images of damage during inspection into the damage analysis model to detect the type of damage, a mask, and two-dimensional judgment criterion information (S410). The damage analysis model is a trained analysis model that, upon input of images of damage during inspection, outputs the type of damage, a mask, and two-dimensional judgment criterion information. Here, the damage mask is a mask used to identify the area where damage exists in the images of damage during inspection, and is typically a two-color image of the same size as the images of damage during inspection. Pixels designated as the first color in the mask indicate the area where the damage exists, and pixels designated as the second color indicate areas other than the area where the damage exists. Two-dimensional judgment criterion information is, as an example, information indicating the presence or absence of corrosion or rust stains near the damage, detected based on the images of damage during inspection.

[0053] Next, the determination unit 3 compares the inspection damage image with the bridge point cloud and uses the inspection damage image, the damage mask, and the bridge point cloud to determine the three-dimensional location of the damage and the type of member to which the damage belongs (S420). The three-dimensional location of the damage typically corresponds to the three-dimensional coordinate value in the point cloud coordinate system of the centroid position of the damage mask which is distributed in a planar manner.

[0054] Next, the determination unit 3 determines the dimensions of the damage (S430). Specifically, the determination unit 3 determines the dimensions of the damage mask in the point cloud coordinate system. The dimensions of the damage are, for example, the length and width dimensions of the rectangular region circumscribing the mask in the point cloud coordinate system. The dimensions of the damage may also be, for example, the diameter of the circle circumscribing the mask in the point cloud coordinate system. The dimensions of the damage may also be evaluated by the area of ​​the mask in the point cloud coordinate system. In addition to the damage mask, the determination unit 3 may also consider the 3D position of the damage when determining the dimensions of the damage. This is because the dimensions of the damage refer to the dimensions of the damage in the real world, and in reality, both the damage mask and the 3D position of the damage (the 3D coordinate values ​​of the damage in the point cloud coordinate system) are necessary to determine the dimensions of the damage. That is, the 3D position of the damage determines the coordinate transformation relationship between the 3D space of the structure and the 2D space of the photograph of the damage. Based on the coordinate transformation relationship, the 2D dimensions of the mask can be transformed into the 3D space of the real world. In fact, measuring the dimensions of damage from a mask requires a coordinate transformation relationship between 3D and 2D space, and the 3D location of the damage can be one method for identifying this coordinate transformation relationship.

[0055] Furthermore, determining the dimensions of damage includes determining the distance between the damage and other damage in its vicinity, in addition to determining the dimensions of the damage itself. Therefore, the dimensions of damage can refer to either the dimensions of the damage itself, or the dimensions of the damage plus the distance between the damage and other damage in its vicinity. In this embodiment, the dimensions of damage refer to the dimensions of the damage plus the distance between the damage and other damage in its vicinity. Note that "other damage" may be of the same type as the damage in question, or it may be of a different type. For example, to determine the extent of a crack, it is preferable to measure the distance to other cracks. On the other hand, to determine the "criteria" for a crack, it is preferable to determine whether there is "free lime" around the crack. As one means of doing so, the distance between the crack and the surrounding free lime can be measured and compared with a threshold value.

[0056] Next, the determination unit 3 determines the degree of damage (S440). Specifically, the determination unit 3 determines the degree of damage based on the type of damage and the dimensions of the damage. For example, if there are two types of damage with the same dimensions, and one type of damage is a crack and the other type of damage is delamination, the determination unit 3 does not determine the degree of damage to be the same, but rather determines the degree of damage using different criteria for each type of damage. That is, a correspondence between the dimensions of the damage and the degree of damage is set for each type of damage. This makes it possible to determine the degree of damage with high accuracy according to the type of damage. In addition, when determining the degree of damage, the determination unit 3 may also consider a damage mask in addition to the type of damage. The mask identifies the area where the damage exists and also identifies the type of damage. That is, a different mask is given for each type of damage.

[0057] To determine the extent of damage, it is advisable to change the criteria for each type of damage. This is because, depending on the type of damage, not only the dimensions of the damage but also its characteristics and other types of damage superimposed on it will influence the assessment of its extent. In this case, it is advisable to consider a damage mask.

[0058] In order to determine the extent of damage, it may be necessary to have characteristics of the damage. For example, in the case of spalling and exposed rebar damage, it is necessary to determine whether only spalling has occurred or whether the rebar is also exposed in order to determine the extent of the damage. In addition, it is also necessary to determine the degree of corrosion of the rebar. As one means of making such determinations, it is preferable to prepare a model that can distinguish between three classes in the mask of spalling and exposed rebar type: spalling, and rebar and corrosion. In this case, a model that has been trained to be able to perform segmentation of these three classes is used.

[0059] To determine the extent of damage, it may be necessary to consider other types of damage superimposed on the damage in question. For example, to determine the extent of water leakage and free lime, it is necessary to determine whether there is only water leakage or also free lime, as well as whether the water leakage or free lime originates from cracks. As one means of making this determination, it is preferable to prepare a model that allows the water leakage and free lime type mask to distinguish between three classes: water leakage, free lime, and cracks. In this case, a model trained to perform segmentation of these three classes is used.

[0060] As another example, it is preferable to consider the width of the crack and its spacing from other cracks when determining the degree of cracking. Deck cracks are a different type of damage from cracks, indicating crack damage in the "deck" of a bridge. In addition to the width of the crack and its spacing from other cracks, the characteristics of the crack (shape or pattern) are also necessary for determining the degree of deck cracking. Figure 8 illustrates the criteria for determining the degree of deck cracking. Figure 8 is reproduced from the Ministry of Land, Infrastructure, Transport and Tourism website. For this reason, one possible method is to determine the characteristics from a mask.

[0061] Furthermore, in parallel with steps S430 and S440, the determination unit 3 determines the criteria for determining damage (S450). Specifically, the determination unit 3 determines the criteria for determining damage by inputting the point cloud of the bridge and its surrounding facilities, the 3D location of the damage, and the 2D criteria information into the criteria diagnosis model. The criteria diagnosis model is a trained diagnostic model that has been trained to output the criteria for determining damage when it is input the point cloud of the bridge and its surrounding facilities, the 3D location of the damage, and the 2D criteria information. The following are examples of specific criteria.

[0062] For example, when the member is a mortar base and the type of damage is delamination, the criteria for judgment include, as an example, "large area" and "progression observed." Also, when the member is pavement and the type of damage is pavement abnormality, the criteria for judgment include, as an example, "accompanied by cross-sectional loss," "concern about impact on the deck," "concern about the spread of impact," "concern about harm to third parties," and "concern about damage to passing vehicles." Also, when the member is a column or wall and the type of damage is delamination, the criteria for judgment include, as an example, "concern about rebar corrosion." Also, when the member is a curb and the type of damage is cracking, the criteria for judgment include, as an example, "accompanied by free lime" and "concern about rebar corrosion." Also, when the member is a deck and the type of damage is water leakage and free lime, the criteria for judgment include, as an example, "concern about impact on other members" and "concern about rebar corrosion." Also, when the member is a main girder and the type of damage is spalling and exposed rebar, the criteria for judgment include, as an example, "concern about rebar corrosion." Furthermore, when the component is a protective fence and the type of damage is delamination, the criteria for judgment may include, for example, "concern about peeling," "concern about impact on third parties," "concern about impact on surface protection work," and "directly below is a road or parking lot." Also, when the component is a breast wall and the type of damage is water leakage / free lime, the criteria for judgment may include, for example, "concern about rebar corrosion" and "progression of the precipitation area."

[0063] Thus, the judgment criteria are additional information for determining the integrity of the damage. The judgment unit 3 determines the judgment criteria based on the point cloud of the bridge and its surrounding facilities and the three-dimensional position of the damage. For example, if there are surrounding facilities such as roads or parking lots where people pass directly below the damage, the judgment criteria will include "concerns about impact on third parties."

[0064] Next, if the three-dimensional position of the damage at the time of inspection matches the three-dimensional position of the damage in the past, the determination unit 3 compares the dimensions of the damage at the time of inspection with the dimensions of the damage in the past to determine whether the damage at the time of inspection is progressive (S460).

[0065] Next, the generation unit 4 generates an inspection report and a 2D damage diagram using the inspection results, which include the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, as well as the point cloud (S470). Next, the generation unit 4 generates 3D data for viewing the results (S480). Simultaneously, the determination unit 3 determines the soundness of the damage by inputting the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage into the diagnostic model (S490).

[0066] The output unit 5 then outputs the inspection report and 2D damage diagram generated by the generation unit 4 as electronic data, and displays the soundness of the damage determined by the judgment unit 3 and the 3D data for viewing results generated by the generation unit 4 on the LCD 1d (S500). By referring to the 3D data for viewing results, understanding of the inspection report and 2D damage diagram is facilitated.

[0067] The embodiments of this disclosure have been described above. The above embodiments have the following features.

[0068] The structural inspection support device 1 includes an acquisition unit 2 (acquisition means) and a determination unit 3 (determination means). The acquisition unit 2 acquires a point cloud (3D data) of a bridge (structure) that is the object of inspection and is composed of multiple members. The acquisition unit 2 acquires inspection damage imaging images taken when bridge damage is detected during bridge inspection. The determination unit 3 classifies the point cloud for each of the multiple members. Based on the point cloud and inspection damage imaging images, the determination unit 3 determines the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage. The determination unit 3 determines the soundness of the damage by inputting the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. With the above configuration, the efficiency of bridge inspection can be improved. In other words, time-consuming and labor-intensive tasks in bridge inspection are automated, and the efficiency and support of inspection are realized. This system will improve the efficiency of bridge inspections, enable accurate diagnoses based on the inspection results, and facilitate decisions regarding early bridge repairs.

[0069] Furthermore, the determination unit 3 determines the dimensions of past damage based on the point cloud and past damage images taken before the inspection. The determination unit 3 determines the progression of the damage at the time of inspection based on the dimensions of the past damage and the dimensions of the damage at the time of inspection that correspond to the past damage. Damage at the time of inspection that corresponds to past damage means damage at the time of inspection that is in the same three-dimensional position as the past damage. The determination unit 3 determines the soundness of the damage by inputting the three-dimensional position of the damage at the time of inspection, the type of member to which the damage at the time of inspection belongs, the type of damage at the time of inspection, the dimensions of the damage at the time of inspection, the degree of the damage at the time of inspection, and the progression of the damage at the time of inspection into the diagnostic model. With the above configuration, the soundness of the damage can be determined with high accuracy by further inputting the progression of the damage at the time of inspection into the diagnostic model.

[0070] Furthermore, the determination unit 3 determines the criteria for determining damage based on the point cloud surrounding the bridge and the three-dimensional location of the damage. The determination unit 3 determines the soundness of the damage by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for determining damage into the diagnostic model. With the above configuration, the soundness of the damage can be determined with high accuracy by further inputting the criteria for determining damage into the diagnostic model.

[0071] Furthermore, the generation unit 4 (generation means) generates an inspection report and a 2D damage diagram using the inspection results, which include the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, as well as a point cloud. With this configuration, human resources and time resources required to create the inspection report and the 2D damage diagram can be saved.

[0072] Furthermore, the generation unit 4 generates 3D data for viewing by superimposing the inspection results, including the 3D location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, onto a point cloud. With this configuration, the inspection results can be easily grasped in 3D.

[0073] Furthermore, the output unit 5 (inspection candidate position output means) calculates the three-dimensional position of past damage based on the point cloud and past damage images taken before the inspection. The output unit 5 displays the three-dimensional position of past damage superimposed on the point cloud as an inspection candidate position. With this configuration, inspectors can easily grasp the inspection candidate positions at the time of inspection by viewing the point cloud.

[0074] Furthermore, determining the dimensions of the damage involves determining the dimensions of the damage and the distance between it and other damage in its vicinity. The determination unit 3 determines the integrity of the damage by inputting the three-dimensional position of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the distance between it and other damage in its vicinity, and the extent of the damage into the diagnostic model. With the above configuration, the integrity of the damage can be determined with high accuracy by further inputting the above-mentioned distance into the diagnostic model. That is, if the distance between two damages that are close to each other is narrow and both damages are cracks, there is a possibility that the two cracks will progress and merge to form one large crack. Similarly, if the two damages are delaminations, there is a possibility that the two delaminations will progress and merge to form one large delamination.

[0075] In the above example, the program can be stored and supplied to the computer using various types of non-transitory computer-readable medium. Non-transitory computer-readable mediums include various types of tangible storage mediums. Examples of non-transitory computer-readable mediums include magnetic storage media, magneto-optical storage media, CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memory. Examples of semiconductor memory include mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory). Alternatively, the program may be supplied to the computer using various types of transient computer-readable mediums. Examples of transient computer-readable mediums include electrical signals, optical signals, and electromagnetic waves. Transitory computer-readable mediums can be supplied to the computer via wired communication channels such as electric wires and optical fibers, or via wireless communication channels.

[0076] Although the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the embodiments described above. Various modifications to the structure and details of the present disclosure can be understood by those skilled in the art within the scope of the present disclosure.

[0077] Each drawing is merely illustrative to illustrate one or more embodiments. Each drawing may be associated with one or more other embodiments rather than with only one specific embodiment. As those skilled in the art will understand, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings, for example, to create embodiments not explicitly shown or described. Not all features or steps shown in any one drawing to illustrate an exemplary embodiment are necessarily required, and some features or steps may be omitted. The order of steps shown in any of the drawings may be changed as appropriate.

[0078] Some or all of the above embodiments may also be described as follows, but are not limited to the following: (Note 1) We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, Structural inspection support system. (Note 2) The structural inspection support system described in Appendix 1, The determination means is, Based on the three-dimensional data and past damage images taken before the inspection, the dimensions of the damage at the time the past damage images were taken are determined. Based on the dimensions of the damage at the time of capturing the previously captured image of the damage, and the dimensions of the damage at the time of capturing the previously captured image of the damage during inspection, the progression of the damage at the time of capturing the previously captured image of the damage during inspection is determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, and the progression of the damage into the diagnostic model. Structural inspection support system. (Note 3) The structural inspection support system described in Appendix 2, The determination means is, Based on the three-dimensional data of the surrounding structure and the three-dimensional location of the damage, the criteria for determining the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage into the diagnostic model. Structural inspection support system. (Note 4) The structural inspection support system described in Appendix 1, An inspection report and a two-dimensional damage diagram are generated using the inspection results, which include the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, along with the three-dimensional data. Further including means for production, Structural inspection support system. (Note 5) The structural inspection support system described in Appendix 1, The following three-dimensional data for viewing is generated by superimposing inspection results, including the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, onto the three-dimensional data. Further including means for production, Structural inspection support system. (Note 6) The structural inspection support system described in Appendix 1, Based on the aforementioned 3D data and past damage images taken prior to the inspection, the 3D position of the damage at the time the past damage images were taken is calculated. The three-dimensional position of the damage at the time of acquisition of the aforementioned past damage image is superimposed on the three-dimensional data and displayed as a candidate inspection position. Further includes means for outputting candidate inspection locations, Structural inspection support system. (Note 7) The structural inspection support system described in Appendix 1, Determining the dimensions of the damage involves determining the dimensions of the damage and the distance between it and other damages in its vicinity. The determination means is, The soundness of the damage is determined by inputting the three-dimensional position of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the distance between the damage and other damages in the vicinity, and the extent of the damage into the diagnostic model. Structural inspection support system. (Note 8) We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, Structural inspection support device. (Note 9) Computers We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Methods for supporting structural inspections. (Note 10) On the computer, We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, A program that makes it function as such.

[0079] Some or all of the elements (e.g., configuration and function) described in Appendices 2 to 7 that are dependent on Appendice 1 may also be dependent on Appendices 8 to 10 in the same manner as those described in Appendices 2 to 7. Some or all of the elements described in any appendice may be applicable to various hardware, software, recording means, systems, and methods for recording software. [Explanation of symbols]

[0080] 1. Structural Inspection Support Device 1a processor 1b Memory 1c communication interface 1d LCD 2 Acquisition part 3 Judgment section 4 Generation part 5. Output section 6 Point cloud DB 7 Image Database

Claims

1. We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, Structural inspection support system.

2. A structural inspection support system according to claim 1, The determination means is, Based on the three-dimensional data and past damage images taken before the inspection, the dimensions of the damage at the time the past damage images were taken are determined. Based on the dimensions of the damage at the time of capturing the previously captured image of the damage, and the dimensions of the damage at the time of capturing the previously captured image of the damage during inspection, the progression of the damage at the time of capturing the previously captured image of the damage during inspection is determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, and the progression of the damage into the diagnostic model. Structural inspection support system.

3. A structural inspection support system according to claim 2, The determination means is, Based on the three-dimensional data of the surrounding structure and the three-dimensional location of the damage, the criteria for determining the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the extent of the damage, the progression of the damage, and the criteria for judging the damage into the diagnostic model. Structural inspection support system.

4. A structural inspection support system according to claim 1, An inspection report and a two-dimensional damage diagram are generated using the inspection results, which include the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, along with the three-dimensional data. Further including means for production, Structural inspection support system.

5. A structural inspection support system according to claim 1, The following three-dimensional data for viewing is generated by superimposing inspection results, including the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage, onto the three-dimensional data. Further including means for production, Structural inspection support system.

6. A structural inspection support system according to claim 1, Based on the aforementioned three-dimensional data and past damage images taken prior to the inspection, the three-dimensional position of the damage at the time the past damage images were taken is calculated. The three-dimensional position of the damage at the time of acquisition of the aforementioned past damage image is superimposed on the three-dimensional data and displayed as a candidate inspection position. Further includes means for outputting candidate inspection locations, Structural inspection support system.

7. A structural inspection support system according to claim 1, Determining the dimensions of the damage involves determining the dimensions of the damage and the distance between it and other damages in its vicinity. The determination means is, The soundness of the damage is determined by inputting the three-dimensional position of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, the distance between the damage and other damages in the vicinity, and the extent of the damage into the diagnostic model. Structural inspection support system.

8. We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, including, Structural inspection support device.

9. Computers We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Methods for supporting structural inspections.

10. On the computer, We acquire 3D data of the structure that is the subject of inspection and is composed of multiple members. During the inspection of the aforementioned structure, an inspection damage image is obtained, capturing images of the damage to the structure. means of acquisition, The three-dimensional data is classified for each of the multiple members, Based on the three-dimensional data and the damage image taken during inspection, the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage are determined. The soundness of the damage is determined by inputting the three-dimensional location of the damage, the type of member to which the damage belongs, the type of damage, the dimensions of the damage, and the extent of the damage into a diagnostic model. Determination means, A program that makes it function as such.

Images