Damaged point extraction device, damaged point extraction method, and damaged point extraction program

Abstract

To provide a damaged point extraction device that can extract and show points where potholes are likely to occur.SOLUTION: A damaged point extraction device includes a target limitation part that narrows down judgment targets in time-series input data based on analysis results of time-series input data that is based on data on reflected microwaves irradiated toward a paved road while driving thereon, and a possible damaged point extraction part that extracts locations on the road where potholes may occur, by using the analysis results of the time-series input data for the judgment targets narrowed down by the target limitation part.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a damaged portion extraction device, a damaged portion extraction method, and a damaged portion extraction program. [Background technology]

[0002] Roads on which vehicles travel daily are subject to damage to the pavement due to the action of the load from the vehicles, temperature changes in the areas where the roads are constructed, and the impact of rainwater on the roads. Specific examples of pavement damage include cracks and potholes. A pothole is a round hole or depression that is formed when the surface layer of the road peels off. In order to quickly grasp such road damage and maintain and manage the roads, the task of grasping the road surface properties is routinely performed, and Patent Document 1 discloses a technology for a trained model generation method and a road surface property determination device that aims to determine the road surface properties at low cost and with high accuracy. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-169705 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology disclosed in Patent Document 1 is capable of determining the presence or absence of road surface conditions such as potholes, but is not capable of extracting and presenting locations where pavement damage is likely to occur in the future.

[0005] The present disclosure has been made in consideration of the above-mentioned points, and aims to provide a damaged area extraction device, a damaged area extraction method, and a damaged area extraction program that are capable of extracting and presenting areas where potholes are likely to occur. [Means for solving the problem]

[0006] A damage area extraction device according to a first aspect of the present disclosure includes a target limitation unit that narrows down judgment targets in time-series input data based on analysis results of time-series input data that is based on data on reflected waves of microwaves irradiated toward a paved road while driving on the road, and a possible damage area extraction unit that extracts locations on the road where potholes may occur, using the analysis results of the time-series input data for the judgment targets narrowed down by the target limitation unit.

[0007] A damaged area extraction device according to a second aspect of the present disclosure is a damaged area extraction device according to the first aspect, wherein the potentially damaged area extraction unit extracts areas on the road where potholes are likely to occur within a predetermined period from the current time.

[0008] A damage location extraction device according to a third aspect of the present disclosure is the damage location extraction device according to the second aspect, wherein the possible damage location extraction unit extracts locations where the analysis results for the time-series input data result in a total of added points set for each of a plurality of analysis items exceed a threshold value as locations on the road where potholes may occur.

[0009] A damaged-point extraction device according to a fourth aspect of the present disclosure is the damaged-point extraction device according to the third aspect, wherein the added points are set for each road linearity of the road.

[0010] A damage location extraction device according to a fifth aspect of the present disclosure is the damage location extraction device according to the third aspect, wherein the points added are the presence or absence of temperature abnormalities in the time series input data, whether the signal strength in the time series input data is within a predetermined range, whether the variance in the time series input data is within a predetermined range, and whether the weighted signal strength in the time series input data, weighted according to the depth from the surface of the road, is within a predetermined range.

[0011] A damage area extraction device according to a sixth aspect of the present disclosure is the damage area extraction device according to the first aspect, wherein the target limitation unit divides the surface of the road into areas of a predetermined size and narrows down the areas to be subjected to judgment in the time-series input data.

[0012] A damage location extraction device according to a seventh aspect of the present disclosure is the damage location extraction device according to the sixth aspect, wherein the target limitation unit narrows down the area to be judged in the time-series input data by using values ​​of accuracy, signal strength, and variance of an abnormal signal in the time-series input data.

[0013] A damage location extraction device according to an eighth aspect of the present disclosure is the damage location extraction device according to the seventh aspect, wherein the target limitation unit determines a range that can cover numerical values ​​in each of the time-series input data, and narrows down the area so that the accuracy, signal strength, and variance values ​​of abnormal signals in the time-series input data fall within the range.

[0014] A damage area extraction device according to a ninth aspect of the present disclosure is the damage area extraction device according to the sixth aspect, wherein the target limitation unit narrows down the area using a plurality of data obtained by slicing the time-series input data in the transverse direction of the road and a plurality of data obtained by slicing the time-series input data in the depth direction of the road.

[0015] A damage area extraction device according to a tenth aspect of the present disclosure is the damage area extraction device according to the ninth aspect, wherein the target limitation unit narrows down the area by combining a first result, which is a prediction result using a plurality of data obtained by slicing the time-series input data in the depth direction of the road, and a second result, which is a prediction result using a plurality of data obtained by slicing the time-series input data in the cross direction of the road.

[0016] A damaged part extraction device according to an eleventh aspect of the present disclosure is the damaged part extraction device according to the first aspect, further comprising a presentation unit that displays the parts extracted by the possibly damaged part extraction unit on a surface image of the road.

[0017] A damage area extraction device according to a twelfth aspect of the present disclosure is the damage area extraction device according to the eleventh aspect, wherein the presentation unit displays the area extracted by the potentially damaged area extraction unit in different colors depending on the period until the occurrence of a pothole.

[0018] A damage area extraction method according to a thirteenth aspect of the present disclosure includes a processor that narrows down judgment targets in time-series input data based on analysis results of time-series input data that is based on data on reflected waves of microwaves irradiated toward a paved road while driving on the road, and performs a process of extracting locations on the road where potholes may occur, using the analysis results of the time-series input data for the narrowed-down judgment targets.

[0019] A damage area extraction program according to a fourteenth aspect of the present disclosure causes a computer to execute a process of narrowing down judgment targets in time-series input data based on analysis results of time-series input data that is based on data on reflected waves of microwaves irradiated toward a paved road while driving on the road, and extracting locations on the road where potholes may occur, using the analysis results of the time-series input data for the narrowed-down judgment targets. Effect of the Invention

[0020] According to the present disclosure, it is possible to provide a damaged point extraction device, a damaged point extraction method, and a damaged point extraction program that are capable of extracting and presenting points where potholes are likely to occur. [Brief description of the drawings]

[0021] [Figure 1] 1 is a diagram showing a schematic configuration of a damage location extraction system including a damage location extraction device according to an embodiment of the present invention. [Diagram 2] 11 is a diagram for explaining detection of a reflected response waveform. FIG. [Diagram 3] 1A and 1B are diagrams illustrating how electromagnetic waves irradiated from an electromagnetic wave device are reflected. [Figure 4]FIG. 2 is a block diagram showing a hardware configuration of the damage point extraction device. [Diagram 5] 2 is a block diagram showing an example of a functional configuration of a damage location extraction device; FIG. [Figure 6] 10 is a flowchart showing a flow of a damage portion extraction process performed by the damage portion extraction device. [Figure 7] FIG. 13 is a diagram showing a state in which a position error occurs for each measurement. [Figure 8] FIG. 13 is a diagram showing image data before alignment processing is performed. [Figure 9] FIG. 13 is a diagram showing image data after a registration process is performed. [Figure 10] FIG. 2 is a diagram illustrating an example of a lane configuration map. [Figure 11] 13 is a flowchart illustrating an example of a narrowing-down process. [Figure 12] FIG. 13 is a graph plotting an example of the relationship between accuracy and variance calculated from reflected wave data. [Figure 13] FIG. 13 is a graph plotting an example of the relationship between intensity and dispersion calculated from reflected wave data. [Figure 14] FIG. 1 is a diagram showing an example of a road alignment. [Figure 15] FIG. 13 is a diagram showing an example of a screen presented by a CPU. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] Hereinafter, an example of an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, the same or equivalent components and parts are given the same reference numerals. In addition, the dimensional ratios of the drawings are exaggerated for the convenience of explanation and may differ from the actual ratios.

[0023] 1 is a diagram showing a schematic configuration of a damage part extraction system including a damage part extraction device according to the present embodiment. As shown in FIG. 1, a damage part extraction system 1 according to the present embodiment is mounted on a vehicle 90.

[0024] The damaged part extraction system 1 is a system that predicts parts of a road 80 paved with asphalt or the like where damage is likely to occur, for example parts where potholes are likely to occur. The damaged part prediction system 1 is configured to include a damaged part extraction device 10 and an electromagnetic wave device 30. The damaged part extraction device 10 is a device that extracts and presents parts of the road 80 where damage is likely to occur, using data obtained by the electromagnetic wave device 30 irradiating the road 80 with electromagnetic waves while a vehicle 90 is traveling.

[0025] The electromagnetic wave device 30 includes a plurality of electromagnetic wave emitting units and receiving units provided on a line. The electromagnetic wave device 30 is provided, for example, on the rear lower part of the vehicle 90 so that the traveling direction of the vehicle 90 is the axial direction and the line direction of the electromagnetic wave device 30 is a direction perpendicular to the axis. The electromagnetic wave emitting units irradiate electromagnetic waves such as microwaves toward the road 80. The receiving units receive the reflected waves reflected at each part inside the pavement of the road 80.

[0026] As the electromagnetic wave irradiation unit, a known electromagnetic wave radar system can be used without any particular limitation, but it is preferable to use a radar system in which a large number of transmitting and receiving sensors are arranged in parallel in terms of work efficiency and accuracy. In addition, it is preferable to use an array antenna, which is an array-shaped antenna, as the transmitting and receiving sensor in terms of work efficiency.

[0027] As shown in Fig. 2, the electromagnetic wave device 30 scans an evaluation target range 95 of a road 80 in the vehicle travel direction, irradiates electromagnetic waves from the surface toward the inside (depth) of the road 80, and receives the reflected waves. This allows the reflected wave intensity to be detected according to the depth for each grid of the evaluation target range 95. One grid is, for example, 1 cm x 1 cm, and one line width can be 2.0 m. In this case, reflected response waveforms for 200 grids are detected for one line.

[0028] The electromagnetic waves emitted from the electromagnetic wave device 30 are reflected at the boundary of the layers of the road 80. However, if there is an abnormality at the boundary of the layers of the road 80 or inside the layers, the electromagnetic waves emitted from the electromagnetic wave device 30 are reflected in a pattern different from that in the normal state. FIG. 3 is a diagram showing the state of reflection of the electromagnetic waves emitted from the electromagnetic wave device 30. When the road 80 is paved with asphalt and the pavement is made up of a surface layer, a base layer, an upper roadbed, and a lower roadbed, if each layer is in a normal state, the electromagnetic waves are reflected as shown in FIG. 3(a). On the other hand, if a crack occurs at the boundary of the layers or a cavity occurs inside the layer, the electromagnetic waves are partially diffusely reflected as shown in FIG. 3(b). The portion where the electromagnetic waves are diffusely reflected as shown in FIG. 3(b) appears normal at first glance in the surface layer, but an abnormality occurs in the lower layer, so it is a portion where the pavement is already damaged or is expected to be damaged in the future.

[0029] When extracting areas where damage is likely to occur on the pavement of the road 80, the damage area extraction device 10 uses data on the electromagnetic waves reflected from the road 80 of the electromagnetic waves irradiated toward the road 80 by the electromagnetic wave device 30. The reflected wave data can be processed in three dimensions and used as image data. When extracting areas where damage is likely to occur on the pavement of the road 80, the damage area extraction device 10 also aligns multiple image data, creates an evaluation grid for the image data, limits the locations to be analyzed, and filters using a point system.

[0030] The electromagnetic wave device 30 outputs information on the acquired reflected response waveform (reflected wave intensity according to depth) for each grid to the storage 20. The damaged part extraction device 10 uses the data recorded in the storage 20 to extract a part where damage is likely to occur in the pavement of the road 80. The electromagnetic wave device 30 is not limited to a form attached to the vehicle 90, and may be in other forms such as a form held by a worker or a form of a handcart. The damaged part extraction device 10 may also take various forms such as a form mounted inside the vehicle 90, a form held by a worker, or a form installed at a place where the extraction work is performed. In the example shown in FIG. 1, the damaged part extraction device 10 is installed outside the vehicle 90 at a place where the extraction work is performed.

[0031] FIG. 4 is a block diagram showing a hardware configuration of the damaged portion extraction device.

[0032] 4, the damaged portion extraction device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, an optical disk drive unit 17, and a communication interface (communication I / F) 18. Each component is connected to each other via a bus 19 so as to be able to communicate with each other.

[0033] The CPU 11 is a central processing unit, and executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14, and executes the program using the RAM 13 as a working area. The CPU 11 controls each of the above components and performs various arithmetic processing according to the program recorded in the ROM 12 or the storage 14. In this embodiment, the ROM 12 or the storage 14 stores a damage location extraction program for extracting locations where damage may occur to the pavement of the road 80.

[0034] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a working area. The storage 14 is configured with an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including an operating system and various data.

[0035] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to perform various input operations. The display unit 16 is, for example, a liquid crystal display, and displays various types of information. The display unit 16 may function as the input unit 15 by adopting a touch panel system.

[0036] The optical disk drive unit 17 reads data stored in various recording media such as a CD-ROM (Compact Disc Read Only Memory) or a Blu-ray disk, and writes data to the recording media.

[0037] The communication interface 18 is an interface for communicating with other devices, and uses standards such as Ethernet (registered trademark), FDDI, and Wi-Fi (registered trademark).

[0038] When executing the above-mentioned damaged portion extraction program, the damaged portion extraction device 10 realizes various functions using the above-mentioned hardware resources. The functional configuration realized by the damaged portion extraction device 10 will be described.

[0039] FIG. 5 is a block diagram showing an example of a functional configuration of the damaged portion extraction device 10. As shown in FIG.

[0040] 5, the damaged part extraction device 10 has, as functional components, an acquisition unit 101, a position alignment unit 102, a target limitation unit 103, a possibly damaged part extraction unit 104, and a presentation unit 105. Each functional component is realized by the CPU 11 reading out and executing a damaged part extraction program stored in the ROM 12 or the storage 14.

[0041] The acquisition unit 101 acquires time-series input data based on data on reflected waves of electromagnetic waves irradiated from the electromagnetic wave device 30 toward the road 80. The time-series input data is data that is a collection of data on reflected waves obtained as a result of the electromagnetic wave device 30 irradiating electromagnetic waves toward the road 80 at different timings. Data included in the time-series input data includes image data generated by three-dimensionally processing the data on the reflected waves of the electromagnetic waves and identification information for linking the image data to the road 80. Furthermore, the data included in the time-series input data may further include information on locations where damage may occur, obtained as a result of applying a trained model to the image data.

[0042] The image data acquired by the acquisition unit 101 may be image data obtained by slicing a three-dimensional image in a predetermined direction. In the following description, the traveling direction of the vehicle 90 on the road 80 is defined as the X direction, the horizontal direction perpendicular to the X direction, i.e., the cross direction of the road 80, is defined as the Y direction, and the depth direction of the road 80 is defined as the Z direction. An image sliced ​​in the Y direction is defined as a Y slice image, and a collection of multiple Y slice images is defined as Y volume data. An image sliced ​​in the Z direction is defined as a Z slice image, and a collection of multiple Z slice images is defined as Z volume data.

[0043] The registration unit 102 performs registration processing to correct position errors for each measurement for the multiple image data acquired by the acquisition unit 101. Details of the registration processing for the multiple image data will be described later.

[0044] The target limiting unit 103 performs a limiting process to limit the location to be analyzed on the image data aligned by the alignment unit 102. By limiting the location to be analyzed by the target limiting unit 103, the damaged part extraction device 10 can extract the damaged part more efficiently than when no limiting process is performed. The limiting process by the target limiting unit 103 will be described in detail later.

[0045] The possibly damaged part extraction unit 104 performs an extraction process of parts where damage may occur on the road 80, using the image data whose target has been limited by the target limiting unit 103. For example, the possibly damaged part extraction unit 104 performs an extraction process of parts where a pothole may occur as damage to the road 80 within a predetermined period from the current time. The extraction process by the possibly damaged part extraction unit 104 will be described later in detail.

[0046] The presentation unit 105 performs a presentation process to visually present the location where damage may occur that is extracted by the possible damage location extraction unit 104. The presentation process by the presentation unit 105 will be described later in detail.

[0047] Next, the operation of the damaged portion extraction device 10 will be described.

[0048] 6 is a flowchart showing the flow of the damaged part extraction process by the damaged part extraction device 10. The CPU 11 reads out a damaged part extraction program from the ROM 12 or the storage 14, loads it into the RAM 13, and executes it to perform the damaged part extraction process.

[0049] In step S11, the CPU 11 acquires time-series input data based on data of reflected waves of electromagnetic waves irradiated from the electromagnetic wave device 30 toward the road 80. A time-series input data acquisition process is executed.

[0050] Following step S11, in step S12, the CPU 11 performs a position alignment process to correct position errors for each measurement for a plurality of image data obtained based on data of reflected waves of electromagnetic waves to the same road 80, which are included in the time-series input data acquired in step S11. Here, a specific example of the position alignment process will be described.

[0051] FIG. 7 is a diagram showing how position errors occur for each measurement. FIG. 7 shows image data obtained in the first measurement of the target section and image data obtained in the subsequent Nth measurement, and shows how position errors occur in each image data. CPU 11 uses the image data of the first measurement of the target section as a reference, and adjusts the position of the image data of the subsequent measurement by using dynamic programming for each lane. CPU 11 associates vertical pixel columns of each image data so that the positions of characteristic locations in the image data match.

[0052] The CPU 11 can deal with positional deviation and expansion / contraction of each image data by performing positional alignment of the image data using dynamic programming.

[0053] The difference between before and after the alignment process by the CPU 11 will be described. Fig. 8 is a diagram showing image data before the alignment process by the CPU 11, and Fig. 9 is a diagram showing image data after the alignment process by the CPU 11. Before the alignment process, the positions of the image data for each measurement are not aligned, but after the alignment process, the positions of the image data for each measurement are aligned. By the CPU 11 performing the alignment process, the damage location extraction device 10 enables analysis in time series at the same location.

[0054] In this embodiment, the image data is aligned using dynamic programming, but the present disclosure is not limited to this example. The damage location extraction device 10 may also align the image data using an image processing technique such as pattern matching.

[0055] Following step S12, in step S13, the CPU 11 performs a narrowing-down process for narrowing down evaluation grids to be judged for the image data of each measurement in which alignment has been performed. Here, a specific example of the narrowing-down process will be described.

[0056] When performing the process of narrowing down the judgment targets, the CPU 11 first divides the image data area corresponding to the road surface into evaluation grid units of a predetermined size. By dividing the image data area into evaluation grid units, each measurement can be analyzed as the same location for each evaluation grid. The size of the evaluation grid can be any size, but as an example, it can be one meter square. The CPU 11 analyzes the data for each evaluation grid. The analysis items are the accuracy of the representative depth, the variance of the signal strength, and the signal strength.

[0057] When dividing the area of ​​the image data in units of evaluation grids, the CPU 11 identifies the evaluation grids by giving numbers indicating the lane, the cross-sectional coordinates within the lane, and the coordinates of the traveling direction. For example, the CPU 11 may use the value of a distance marker (kilometer post) used for road management as the number indicating the coordinate of the traveling direction. In addition, when the lane is composed of multiple lanes, the cross-sectional coordinates within the lane may be set to 0 for a specific lane, positive for the center divider, and negative for the sidewalk. Information on the road gradient may be used as information for linking with the distance marker. Linking each evaluation grid with the value of a distance marker (kilometer post) used for road management makes it possible to use the analysis of road management information and information for identifying the local position of the analysis results.

[0058] When identifying each evaluation grid, the damage location extraction device 10 is operated, and a lane configuration map of the route section to be managed is created in advance. The lane configuration map is created because it is necessary to associate lanes in the same block in the time series direction during positioning, and to narrow down the search to the same lanes on the same route in the time series analysis.

[0059] FIG. 10 is a diagram showing an example of a lane configuration map. As shown in FIG. 10, the lane configuration map introduces a global lane ID that is unified throughout the entire route, and adds the number of lanes in each block and the global lane ID of lane 1 to the input data. By registering the lane offset from lane 1 in each block, lane numbers in the same block can be unified even if the driving lane cannot be measured. In addition, by registering the global lane ID of lane 1, the same lane on the same route can be distinguished. In addition, as shown in FIG. 10, even if there is section data with a different number of lanes depending on the lane configuration, or even if it is necessary to identify lanes that could not be measured at the time of measurement due to lane regulations, etc., it is possible to distinguish the same lane on the route by introducing the lane configuration map shown in FIG. 10.

[0060] In addition, as location-specific information on a digital map or GIS (Geographic Information System), latitude and longitude information from a GNSS (Global Navigation Satellite System) may be linked to an evaluation grid at the center of each lane in the transverse direction.

[0061] The CPU 11 narrows down the analysis targets by filtering the image data to which the lane configuration map has been applied according to narrowing conditions, etc. During the narrowing down process, the CPU 11 performs the narrowing down process in a state where the values ​​of mileposts (kilometer posts) that manage the roads are linked to data such as the coordinates of the evaluation grid and the road structure (gradient).

[0062] Fig. 11 is a flow chart showing an example of the narrowing down process by the CPU 11. As shown in Fig. 11, the narrowing down condition items include, for example, the joint probability, overlapping grids, the position of the rut, the extraction of abnormal signals multiple times, and the accuracy, strength, and variance values ​​of the abnormal signals. The joint probability is calculated by (the accuracy of the Z volume data x the accuracy of the Y volume data). 1 / 2 The evaluation grid is narrowed down based on the value obtained by, for example, extracting evaluation grids whose probability of abnormality is greater than a predetermined value (for example, 0.5 or more). The accuracy of Y volume data is an example of a first result of the present disclosure. The accuracy of Z volume data is an example of a second result of the present disclosure. The overlapping grid is narrowed down by eliminating overlapping counts of the same evaluation grid in a time series. The position of the rut is narrowed down by the evaluation grid at the position of the rut where the tires of the vehicle are thought to pass through frequently. The multiple extraction is narrowed down by the evaluation grid at the location where the abnormal signal is detected multiple times. The accuracy, intensity, and variance values ​​are narrowed down by the evaluation grid based on the values ​​calculated from the data of the reflected wave, and the numerical range to be narrowed down is specified based on the numerical trend corresponding to the location where the pothole occurs from the value indicating the relationship between the two axes, and the evaluation grid is narrowed down by extracting the location that meets the condition.

[0063] In step S101, the CPU 11 narrows down the evaluation grids by joint probability. After narrowing down the evaluation grids by joint probability, the CPU 11 narrows down the evaluation grids by overlapping grids in step S102. After narrowing down the evaluation grids by overlapping grids, the CPU 11 narrows down the evaluation grids by rut positions in step S103. After narrowing down the evaluation grids by rut positions, the CPU 11 narrows down the evaluation grids by locations extracted multiple times in step S104. After narrowing down the evaluation grids by locations extracted multiple times, the CPU 11 narrows down the evaluation grids by accuracy, strength, and variance values ​​in step S105.

[0064] Although the order of narrowing down can be changed, it is desirable to narrow down the evaluation grids based on overlapping grids at an early stage of the narrowing down process, and to narrow down the evaluation grids based on the accuracy, strength, and variance values ​​at the final stage of the narrowing down process.

[0065] Here is an example of narrowing down the search results using the values ​​of accuracy, strength, and variance. As mentioned above, narrowing down the search results using the values ​​that indicate the relationship between two axes specifies the range of values ​​to narrow down based on the numerical trend corresponding to the pothole occurrence locations. The following example shows the case where the relationship between variance and strength, and the relationship between variance and accuracy are used.

[0066] FIG. 12 is a graph showing an example of the relationship between the accuracy and the variance calculated from the data of the reflected wave, and FIG. 13 is a graph showing an example of the relationship between the intensity and the variance calculated from the data of the reflected wave. The graphs in FIG. 12 and FIG. 13 show examples of the relationship between the accuracy and the variance and the relationship between the intensity and the variance for 12 evaluation grids where abnormal signals were detected before the pothole occurred at the pothole occurrence point on the road. In other words, the graphs in FIG. 12 and FIG. 13 are graphs plotting time series data before the pothole occurrence at 12 points where the pothole actually occurred. The damage point extraction device 10 according to the present embodiment finds a range in each graph that can cover the values ​​of the 12 time series input data. Specifically, the damage point extraction device 10 according to the present embodiment finds an area in the graphs in FIG. 12 and FIG. 13 that can cover the signals at 12 points. The area indicated by the symbol A1 in FIG. 12 and the area indicated by the symbol A2 in FIG. 13 are areas in which the signals at 12 points are covered in each graph. In other words, the area indicated by symbol A1 in Fig. 12 and the area indicated by symbol A2 in Fig. 13 are areas in which all of the signals from 12 points fall in each graph. The damage location extraction device 10 narrows down the evaluation grid based on the relationship between the accuracy and variance and the relationship between the strength and variance in the obtained range, thereby narrowing down the evaluation grid using the relationship between the variance and the strength and the relationship between the variance and the accuracy.

[0067] 12 and 13, the CPU 11 specifies the narrowed-down numerical range by plotting the relationship between the accuracy, intensity, and variance in a two-dimensional plane, but the present disclosure is not limited to such an example. The CPU 11 may plot the relationship between the accuracy, intensity, and variance in a three-dimensional space to specify the narrowed-down numerical range.

[0068] Following step S13, in step S14, the CPU 11 executes an extraction process for extracting locations with a high probability of pothole occurrence from the evaluation grids that have been subjected to the narrowing process. Here, the extraction process will be described in detail.

[0069] As an extraction process, the CPU 11 extracts abnormal points for each road linearity by filtering. Here, the road linearity includes the gradient in the transverse direction (transverse gradient) and the gradient in the longitudinal direction (longitudinal gradient). FIG. 14 is a diagram showing an example of a road linearity. FIG. 14 (A) shows the transverse gradient of the road, and FIG. 14 (B) shows the longitudinal gradient. In this embodiment, the CPU 11 extracts abnormal points for each of 36 road linearities, including six transverse gradients 1 to 6 and six longitudinal gradients A to F. In this embodiment, the filtering method used is a point-adding system rather than a narrowing-down system, in order to extract points that meet multiple conditions.

[0070] In this embodiment, infrared anomaly, variance, intensity, and depth weighting are adopted as the point-adding elements. Infrared anomaly is an element that judges the presence or absence of a temperature anomaly by an infrared image, and adds points if there is a temperature anomaly. Variance is an element that adds points if the variance value is within a certain threshold. Intensity is an element that adds points if the intensity value is within a certain threshold. Depth weighting is an element that adds points if the depth weighting value is within a certain threshold. Here, the depth weighting refers to a value obtained by weighting the signal strength as the depth becomes shallower, so that the weighting for shallow signals becomes larger, and is an example of the weighted signal strength of the present disclosure. The reason why the signal strength is weighted as the depth becomes shallower in this way is based on the assumption that signal changes at shallow points from the surface increase the risk of pothole occurrence. The threshold value of each element can be determined from the numerical tendency of actual pothole occurrence. And the threshold value can be determined from the actual value of the pothole occurrence point in the past for each road gradient division. In this embodiment, a predetermined value is given for each point-adding element for each evaluation grid if the condition is met. As an example, five points will be awarded for each bonus element if the conditions are met.

[0071] Then, the CPU 11 selects an evaluation grid where the total value of the points for each point-adding element is equal to or greater than a predetermined threshold as an evaluation grid where a pothole may occur. For example, the CPU 11 selects an evaluation grid where the total value of the points is equal to or greater than 15 as a grid where a pothole may occur.

[0072] Following step S14, in step S15, the CPU 11 executes a presentation process for presenting the extracted locations with a high probability of pothole occurrence. Here, the presentation process will be described in detail.

[0073] Fig. 15 is a diagram showing an example of a screen presented by the CPU 11. Fig. 15 shows an example of a screen in which a surface image, Y slice data, and Z slice data of a road are displayed on one screen, and locations where a pothole is likely to occur are presented. The screen shown in Fig. 15 shows that an evaluation grid indicated by reference numeral 21 is a location selected as an evaluation grid where a pothole is likely to occur.

[0074] When presenting locations where a pothole is likely to occur, the CPU 11 may present them in different colors depending on the time until the pothole occurs. For example, the CPU 11 may present locations where a pothole is likely to occur in blue if the time until the pothole occurs is 4 to 6 months, in yellow if the time until the pothole occurs is 2 to 4 months, and in red if the time until the pothole occurs is less than 2 months. By presenting locations where a pothole is likely to occur in different colors depending on the time until the pothole occurs, the damaged portion extraction device 10 can indicate locations that require early repair.

[0075] By executing the presentation process in this manner, the damaged portion extraction device 10 can visually present the portions where the possibility of pothole occurrence is high.

[0076] As described above, according to the embodiment of the present disclosure, it is possible to provide a damaged point extraction device 10 that extracts points where a pothole is highly likely to occur.

[0077] Although the embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, the technical scope of the present disclosure is not limited to such examples. It is clear that a person having ordinary knowledge in the technical field of the present disclosure can conceive of various modified examples or examples of modifications within the scope of the technical ideas described in the claims, and it is understood that these modified examples or examples of modifications also belong to the technical scope of the present disclosure.

[0078] In addition, the effects described in the above embodiments are explanatory or exemplary, and are not limited to those described in the above embodiments. In other words, the technology according to the present disclosure may achieve other effects that are obvious to a person having ordinary skill in the technical field of the present disclosure from the description in the above embodiments, in addition to or in place of the effects described in the above embodiments.

[0079] In addition, the damaged portion extraction process executed by the CPU by reading the software (program) in each of the above embodiments may be executed by various processors other than the CPU. In this case, examples of the processor include a PLD (Programmable Logic Device) such as an FPGA (Field-Programmable Gate Array) whose circuit configuration can be changed after manufacture, and a dedicated electric circuit such as an ASIC (Application Specific Integrated Circuit) which is a processor having a circuit configuration designed exclusively for executing a specific process. In addition, the damaged portion extraction process may be executed by one of these various processors, or may be executed by a combination of two or more processors of the same or different types (for example, a plurality of FPGAs, and a combination of a CPU and an FPGA). In addition, the hardware structure of these various processors is, more specifically, an electric circuit in which circuit elements such as semiconductor elements are combined.

[0080] In addition, in each of the above embodiments, the damaged portion extraction processing program is stored (installed) in advance in a ROM or storage, but the present invention is not limited to this. The program may be provided in a form recorded on a non-transitory recording medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The program may also be downloaded from an external device via a network. [Explanation of symbols]

[0081] 10 Damage extraction device 30 Electromagnetic wave device 80 road 90 vehicles

Claims

1. an object limiting unit that narrows down objects to be determined in the time-series input data based on an analysis result of the time-series input data based on data of reflected waves of microwaves irradiated toward a paved road while the vehicle is traveling on the road; a possible damage location extraction unit that extracts locations on the road where potholes may occur using an analysis result of the time-series input data for the determination targets narrowed down by the target limitation unit; and A damage location extraction device comprising:

2. The damaged portion extraction device according to claim 1 , wherein the possibly damaged portion extraction unit extracts portions on the road where a pothole is likely to occur within a predetermined period from a current time.

3. 3. The damage location extraction device according to claim 2, wherein the possible damage location extraction unit extracts locations where a total of added points set for each of a plurality of analysis items in the analysis results for the time-series input data exceeds a threshold value as locations where a pothole is likely to occur on the road within the specified period.

4. The damaged portion extraction device according to claim 3 , wherein the added points are set for each road alignment of the road.

5. 4. The damage location extraction device according to claim 3, wherein the points added are whether or not there is a temperature abnormality in the time series input data, whether the signal strength in the time series input data is within a predetermined range, whether the variance in the time series input data is within a predetermined range, and whether a weighted signal strength in the time series input data weighted according to a depth from the surface of the road is within a predetermined range.

6. 2. The damaged portion extraction device according to claim 1, wherein the target limiting unit divides the surface of the road into regions of a predetermined size and then narrows down the regions to be subjected to judgment in the time-series input data.

7. 7. The damaged portion extraction device according to claim 6, wherein the target limiting unit narrows down the area to be determined in the time-series input data by using values ​​of probability, signal strength, and variance of an abnormal signal in the time-series input data.

8. The damage location extraction device according to claim 7, wherein the target limitation unit determines a range that can be covered by values ​​in each of the time-series input data, and narrows down the area so that the accuracy, signal strength, and variance values ​​of the abnormal signal in the time-series input data fall within the range.

9. The damage location extraction device according to claim 6 , wherein the target limitation unit narrows down the area using a plurality of data obtained by slicing the time-series input data in a transverse direction of the road and a plurality of data obtained by slicing the time-series input data in a depth direction of the road.

10. 10. The damage location extraction device according to claim 9, wherein the target limitation unit narrows down the area by combining a first result, which is a prediction result using a plurality of data obtained by slicing the time-series input data in a depth direction of the road, and a second result, which is a prediction result using a plurality of data obtained by slicing the time-series input data in a cross direction of the road.

11. The damage portion extraction device according to claim 1 , further comprising a presentation unit that displays the portion extracted by the possibly damaged portion extraction unit on the surface image of the road.

12. The damaged portion extraction device according to claim 11 , wherein the presentation unit displays the portion extracted by the possibly damaged portion extraction unit in a different color depending on a period until a pothole occurs.

13. The processor: narrowing down judgment targets in the time-series input data based on an analysis result of time-series input data based on data of reflected waves of microwaves irradiated toward a paved road while the vehicle is traveling on the road; Using the analysis results of the time-series input data for the narrowed down judgment targets, locations on the road where potholes are likely to occur are extracted. A damage extraction method for performing processing.

14. On the computer, narrowing down judgment targets in the time-series input data based on an analysis result of time-series input data based on data of reflected waves of microwaves irradiated toward a paved road while the vehicle is traveling on the road; Using the analysis results of the time-series input data for the narrowed down judgment targets, locations on the road where potholes are likely to occur are extracted. A damage extraction program that executes the process.

Images