A method for identifying crack width of asphalt pavement based on radar amplitude characteristics

Abstract

The invention discloses a crack width identification method of radar amplitude characteristics of asphalt pavement, which relates to the technical field of pavement crack identification, including collecting radar data of each detection channel, analyzing the longitude and latitude information of all radar data points, and obtaining a complete radar projection map of the asphalt pavement; identifying the edge of each crack, analyzing the crack skeleton curve, calculating the minimum distance from each crack edge point to the skeleton curve, and obtaining the first average crack width of each crack; obtaining the reflection amplitude of each internal point of each crack and the reflection amplitude mean of the data point at the disease-free place, calculating the average difference coefficient of the crack amplitude, and building a crack width analysis model; marking the crack skeleton point, calculating the amplitude difference coefficient of the crack skeleton point, importing the amplitude difference coefficient of the crack skeleton point into the crack width analysis model, and obtaining the crack width at each crack skeleton point of the crack. It provides a scientific basis for repair work and improves the pertinence and effectiveness of repair work.

Description

Technical Field

[0001] The invention relates to the technical field of pavement crack identification, and in particular to a method for identifying crack width of an asphalt pavement using radar amplitude characteristics. Background Art

[0002] Under the influence of driving load, temperature, ultraviolet rays, etc., the road pavement structure will inevitably suffer damage as the operation time increases, such as cracks, subsidence, and interlayer voids. Quickly, effectively, and accurately find the location and type of pavement structure diseases, determine their severity, and predict the development of diseases, which is of great significance for scientifically and rationally determining the maintenance timing and formulating maintenance plans. Among the various types of pavement structure diseases, crack diseases are the most numerous and have a great impact on the service performance of pavement structures. Since the spatial span of crack diseases is large and the features usually span multiple lanes, in the interpretation of 3D ground penetrating radar crack disease images, single-channel 3D ground penetrating radar data is currently widely used to identify crack diseases, which has the problems of narrow auxiliary judgment area, high identification difficulty, and certain misjudgment.

[0003] In a Chinese invention application with application publication number CN112666554A, a method for identifying crack width of an asphalt pavement using radar amplitude characteristics is disclosed, comprising the following steps: step 1, collecting three-dimensional ground penetrating radar data and determining the crack pile number; step 2, performing time zero point, filtering, gain, and offset imaging processing on the three-dimensional ground penetrating radar data to highlight abnormal positions in the three-dimensional radar image; step 3, calculating the average amplitude value A1 of the defect-free area in the filtered three-dimensional ground penetrating radar data; step 4, extracting the amplitude value A2 of the echo signal of the three-dimensional ground penetrating radar at the crack; step 5, performing feature extraction and identification on the average amplitude difference value Ad, and calculating the corresponding relationship between the average amplitude difference value Ad and the crack width d.

[0004] In the above invention application, the feature extraction and identification of the average amplitude difference value Ad are performed by analyzing the three-dimensional ground penetrating radar data, and the corresponding relationship between the average amplitude difference value Ad and the crack width d is calculated. However, the single-channel data is still analyzed to identify the crack disease in the data, and other data in the same section are not fully considered. When the crack length is long and spans multiple channels, the single-channel three-dimensional ground penetrating radar is used to identify the crack disease, which has the problems of narrow auxiliary determination area, high identification difficulty and inaccurate crack area determination. Moreover, the identified cracks cannot correspond to the specific location of the road surface, and it takes time to conduct field surveys during maintenance to find the actual location of the cracks.

[0005] To this end, the present invention provides a method for identifying crack width of asphalt pavement using radar amplitude characteristics. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the shortcomings of the prior art, the present invention provides a method for identifying crack width of asphalt pavement using radar amplitude characteristics. The present invention calculates the longitude and latitude coordinates of radar points, maps radar data onto satellite maps, combines multi-channel radar data to form a complete radar projection map of the asphalt pavement, identifies the edge of each crack in the radar projection map of the asphalt pavement, analyzes the crack skeleton curve, and obtains the first average crack width of each crack based on the image analysis results. Dp , build a crack width analysis model, and analyze the crack width at each crack skeleton point based on the crack width analysis model, which provides a scientific basis for the repair work, helps to determine the repair scope, repair materials and repair process, and improves the pertinence and effectiveness of the repair work, thereby solving the technical problems recorded in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above purpose, the present invention is implemented by the following technical scheme: a method for identifying crack width of asphalt pavement radar amplitude characteristics, comprising the following steps:

[0010] Divide the asphalt pavement into multiple detection lanes and determine the center line of each detection lane, collect radar data from each detection lane, analyze the longitude and latitude information of all radar data points, project the radar data points onto the satellite map, and obtain a complete radar projection map of the asphalt pavement;

[0011] Identify the edge of each crack in the radar projection image of the asphalt pavement, analyze the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack, and calculate the minimum distance from each crack edge point to the skeleton curve , after statistical sorting, the first average width of each crack is obtained Dp ;

[0012] Determine the boundary range of the crack, check the position of each radar point, mark it as the internal point of the crack or the data point without damage, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj , build a crack width analysis model;

[0013] Check the position of each radar point, mark the crack skeleton point, and obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point , the crack skeleton point amplitude difference coefficient Import the crack width analysis model to obtain the crack width at each crack skeleton point.

[0014] Furthermore, the asphalt pavement is divided into multiple inspection lanes according to the signs and markings on the road and the center line of each inspection lane is determined. An inspection vehicle equipped with a three-dimensional ground penetrating radar is used to travel on the center lines of multiple inspection lanes to collect radar data of each inspection lane, and a radar XYZ three-dimensional coordinate system is established, with the direction of travel of the radar antenna as the X-axis, the depth direction perpendicular to the earth as the Z-axis, and the direction perpendicular to the XZ plane as the Y-axis, to obtain XY slices, YZ slices, and XZ slices.

[0015] Since the crack signal is not obvious on the YZ plane and XZ plane, it is more obvious on the XY plane. This is related to the fact that the crack disease is long, but the width and height are generally small. Therefore, the XY plane of the 3D radar data is mainly intercepted to identify the crack disease.

[0016] Furthermore, the recorded position information of the navigation file of the inspection vehicle is obtained, and the radar data points on the x-axis are matched according to the recorded time to obtain the radar data points on the center line of the inspection road. The longitude and latitude data is recorded as the longitude and latitude coordinates of the center line radar point The remaining radar data points not on the x-axis are matched with the centerline radar points according to the acquisition time and recorded as radar points to be estimated. The longitude and latitude coordinates of the radar point to be estimated are .

[0017] Further, based on the latitude and longitude coordinates of the centerline radar point Calculate the latitude and longitude coordinates of each radar point to be estimated :

[0018]

[0019] in, Angle is a straight line and the angle between due south, , is the average radius of the Earth, represents the magnitude of a vector, i.e. is the centerline radar point and radar points to be estimated The distance between.

[0020] Furthermore, the complete asphalt pavement radar projection image is imported into an edge detection algorithm (such as Canny edge detection) to identify the edge of each crack in the asphalt pavement radar projection image, and then the detected crack edges are separated from the entire asphalt pavement radar projection image through a binarization operation to obtain the edge image of each crack and number them sequentially.

[0021] Furthermore, the edge image of the crack is refined to obtain a single-pixel wide crack skeleton curve, and endpoints and inflection points are extracted as feature points on the crack skeleton curve and the edge image of the crack. The feature points on the crack edge image are matched with the feature points on the crack skeleton curve using a feature point matching algorithm, and the crack skeleton curve is mapped to the edge image of the crack.

[0022] Furthermore, each crack edge point and all points on the skeleton curve are traversed, and the Euclidean distance between all points on the skeleton curve and a certain crack edge point is calculated to obtain the minimum distance from each crack edge point to the skeleton curve. , after statistical sorting, the first average width of each crack is obtained Dp , the first average width of the crack Dp For 2 ;

[0023] in, c is the sequential numbering of the crack edge points of the same crack, The minimum distance from all crack edge points to the skeleton curve of the same crack The mean of .

[0024] Furthermore, after identifying the edge of each crack in the radar projection image of the asphalt pavement, the boundary range of the crack is determined, and the position of each radar point is checked to determine whether it is within the boundary range of the crack. If the radar point is within the boundary range of the crack, it is marked as an internal point of the crack, and its coordinates and radar data are extracted to obtain the reflection amplitude of each internal point of each crack. The remaining radar points that are not marked as internal points of the cracks are marked as data points without defects. The reflection amplitudes of all data points without defects are extracted, and the mean reflection amplitudes of the data points without defects are obtained after statistics. .

[0025] Furthermore, the reflection amplitude of each internal point of each crack is obtained. and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj :

[0026]

[0027] in, i Indicates the sequential number of the points inside the crack, .

[0028] Furthermore, the average difference coefficient of crack amplitude of multiple cracks is obtained. Cj and the first average width of cracks Dp, select part of the data as sample data, build a crack width analysis model based on the linear regression algorithm, and output the crack width analysis model after sample data training and testing.

[0029] Furthermore, a fusion image of the crack skeleton curve mapped onto the edge image of the crack is obtained, and the position of each radar point is checked to determine whether it is located on the crack skeleton curve. If the radar point is on the crack skeleton curve, it is marked as the skeleton point of the crack, and its coordinates and radar data are extracted to obtain the reflection amplitude of each crack skeleton point. .

[0030] Furthermore, the reflection amplitude of each crack skeleton point is obtained and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point :

[0031]

[0032] in, j Indicates the latitude and longitude position of the crack skeleton point;

[0033] The amplitude difference coefficient of the crack skeleton point Import the crack width analysis model to obtain the crack width at each crack skeleton point.

[0034] (III) Beneficial effects

[0035] The present invention provides a method for identifying crack width of asphalt pavement using radar amplitude characteristics, which has the following beneficial effects:

[0036] 1. Divide the asphalt pavement into multiple test lanes and determine the center line of each test lane, collect radar data from each test lane, analyze the latitude and longitude information of all radar data points, project the radar data points onto the satellite map, and obtain a complete radar projection map of the asphalt pavement. The pavement structure can be intuitively seen, making the test results more intuitive and easy to understand. During the maintenance and repair process, the radar projection map can be used to quickly locate the problem area and improve the repair efficiency and quality.

[0037] 2. Identify the edge of each crack in the radar projection image of the asphalt pavement, analyze the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack, and calculate the minimum distance from each crack edge point to the skeleton curve , after statistical sorting, the first average width of each crack is obtained Dp , the morphology and distribution of cracks can be intuitively understood, which helps to identify the expansion trend of cracks and potential risk areas.

[0038] 3. Determine the boundary range of the crack, check the position of each radar point, mark it as the internal point of the crack or the data point without damage, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj , building a crack width analysis model, can accurately evaluate the state of cracks, improve detection efficiency, guide repair work, predict crack development trends, and standardize and repeatable analysis.

[0039] 4. Check the position of each radar point, mark the crack skeleton point, and obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point , the crack skeleton point amplitude difference coefficient The crack width analysis model is imported to obtain the crack width at each crack skeleton point, which provides a scientific basis for the repair work and helps to determine the repair scope, repair materials and repair process, thereby improving the pertinence and effectiveness of the repair work. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of a method for identifying crack width of an asphalt pavement using radar amplitude characteristics according to the present invention;

[0041] Figure 2 A schematic diagram of the arrangement of the three-dimensional ground penetrating radar survey track of the present invention;

[0042] Figure 3 It is a schematic diagram of the processing of the radar projection image of the asphalt pavement according to the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figures 1 to 3 The present invention provides a method for identifying crack width of asphalt pavement using radar amplitude characteristics, comprising the following steps:

[0045] Step 1: Divide the asphalt pavement into multiple detection lanes and determine the center line of each detection lane, collect radar data from each detection lane, analyze the longitude and latitude information of all radar data points, project the radar data points onto a satellite map, and obtain a complete radar projection map of the asphalt pavement.

[0046] The step 1 includes the following contents:

[0047] Step 101: Divide the asphalt road surface into multiple test lanes according to the markings on the road and determine the center line of each test lane (see Figure 2 ), use an inspection vehicle equipped with a three-dimensional ground penetrating radar to drive on the center lines of multiple inspection lanes to collect radar data from each inspection lane, and establish a radar XYZ three-dimensional coordinate system, with the direction of radar antenna travel as the X-axis, the depth direction perpendicular to the earth as the Z-axis, and the direction perpendicular to the XZ plane as the Y-axis to obtain XY slices, YZ slices, and XZ slices.

[0048] Since the crack signal is not obvious on the YZ plane and XZ plane, it is more obvious on the XY plane. This is related to the fact that the crack disease is long, but the width and height are generally small. Therefore, the XY plane of the 3D radar data is mainly intercepted to identify the crack disease.

[0049] Step 102: Obtain the recorded position information of the navigation file of the inspection vehicle, match the recorded time with the radar data point on the x-axis, and obtain the radar data point on the center line of the inspection road. The longitude and latitude data is recorded as the longitude and latitude coordinates of the center line radar point The remaining radar data points not on the x-axis are matched with the centerline radar points according to the acquisition time and recorded as radar points to be estimated. The longitude and latitude coordinates of the radar point to be estimated are .

[0050] in, a represents the acquisition time of radar data, b represents the sequence number of radar points to be estimated at the same acquisition time, .

[0051] Step 103: Assuming that the ground plane is approximately the Euclidean plane, based on the latitude and longitude coordinates of the centerline radar point Calculate the latitude and longitude coordinates of each radar point to be estimated :

[0052]

[0053] in, Angle is a straight line and the angle between due south, , is the average radius of the Earth, represents the magnitude of a vector, i.e. is the centerline radar point and radar points to be estimated The distance between.

[0054] Step 104: project the radar data points onto a satellite map based on the latitude and longitude information of all radar data points to obtain a complete radar projection map of the asphalt road surface.

[0055] When using, combine the contents in steps 101 to 104:

[0056] The asphalt pavement is divided into multiple test lanes and the center line of each test lane is determined. The radar data of each test lane is collected, the longitude and latitude information of all radar data points are analyzed, and the radar data points are projected onto the satellite map to obtain a complete radar projection map of the asphalt pavement. The pavement structure can be intuitively seen, making the test results more intuitive and easy to understand. During the maintenance and repair process, the radar projection map can be used to quickly locate the problem area and improve the repair efficiency and quality.

[0057] Step 2: Identify the edge of each crack in the radar projection image of the asphalt pavement, analyze the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack, and calculate the minimum distance from each crack edge point to the skeleton curve. , after statistical sorting, the first average width of each crack is obtained Dp.

[0058] The step 2 includes the following contents:

[0059] Step 201: import the complete asphalt pavement radar projection image into an edge detection algorithm (such as Canny edge detection), identify the edge of each crack in the asphalt pavement radar projection image, and then separate the detected crack edges from the complete asphalt pavement radar projection image through a binarization operation to obtain an edge image of each crack and number them sequentially.

[0060] Step 202: Perform a thinning operation on the crack edge image to obtain a crack skeleton curve with a single pixel width, extract endpoints and inflection points on the crack skeleton curve and the crack edge image as feature points, use a feature point matching algorithm to match the feature points on the crack edge image with the feature points on the crack skeleton curve, and map the crack skeleton curve to the crack edge image (see Figure 3 ).

[0061] Step 203: traverse each crack edge point and all points on the skeleton curve, calculate the Euclidean distance between all points on the skeleton curve and a certain crack edge point, and obtain the minimum distance from each crack edge point to the skeleton curve. , after statistical sorting, the first average width of each crack is obtained Dp , the first average width of the crack Dp For 2 .

[0062] in, c is the sequential numbering of the crack edge points of the same crack, The minimum distance from all crack edge points to the skeleton curve of the same crack The mean of .

[0063] When using, combine the contents in steps 201 to 203:

[0064] Identify the edge of each crack in the radar projection image of the asphalt pavement, analyze the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack, and calculate the minimum distance from each crack edge point to the skeleton curve , after statistical sorting, the first average width of each crack is obtained Dp , the morphology and distribution of cracks can be intuitively understood, which helps to identify the expansion trend of cracks and potential risk areas.

[0065] Step 3: Determine the boundary range of the crack, check the position of each radar point, mark it as the internal point of the crack or the data point without damage, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack. and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj , build a crack width analysis model.

[0066] The step three includes the following contents:

[0067] Step 301: After identifying the edge of each crack in the radar projection image of the asphalt pavement, determine the boundary range of the crack, check the position of each radar point, and determine whether it is within the boundary range of the crack. If the radar point is within the boundary range of the crack, mark it as an internal point of the crack, and extract its coordinates and related radar data to obtain the reflection amplitude of each internal point of each crack. The remaining radar points that are not marked as internal points of the cracks are recorded as data points without defects. The reflection amplitudes of all data points without defects are extracted, and the mean reflection amplitude of the data points without defects is obtained after statistics. .

[0068] Step 302: Obtain the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj :

[0069]

[0070] in, i Indicates the sequential number of the points inside the crack, .

[0071] Step 303: Obtain the average difference coefficient of crack amplitudes of multiple cracks Cj and the first average width of cracks Dp , select part of the data as sample data, build a crack width analysis model based on the linear regression algorithm, and output the crack width analysis model after sample data training and testing.

[0072] When using, combine the contents in steps 301 to 303:

[0073] Determine the boundary range of the crack, check the position of each radar point, mark it as the internal point of the crack or the data point without damage, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj , building a crack width analysis model, can accurately evaluate the state of cracks, improve detection efficiency, guide repair work, predict crack development trends, and standardize and repeatable analysis.

[0074] Step 4: Check the position of each radar point, mark the crack skeleton point, and obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point , the crack skeleton point amplitude difference coefficient Import the crack width analysis model to obtain the crack width at each crack skeleton point.

[0075] The step 4 includes the following contents:

[0076] Step 401: Obtain a fused image of the crack skeleton curve mapped onto the edge image of the crack, check the position of each radar point, and determine whether it is located on the crack skeleton curve. If the radar point is on the crack skeleton curve, mark it as the skeleton point of the crack, and extract its coordinates and related radar data to obtain the reflection amplitude of each crack skeleton point. .

[0077] Step 402: Obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point :

[0078]

[0079] in, j Indicates the latitude and longitude position of the crack skeleton point.

[0080] The amplitude difference coefficient of the crack skeleton point Import the crack width analysis model to obtain the crack width at each crack skeleton point.

[0081] When using, combine the contents in steps 401 and 402:

[0082] Check the position of each radar point, mark the crack skeleton point, and obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point , the crack skeleton point amplitude difference coefficient The crack width analysis model is imported to obtain the crack width at each crack skeleton point, which provides a scientific basis for the repair work and helps to determine the repair scope, repair materials and repair process, thereby improving the pertinence and effectiveness of the repair work.

[0083] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0084] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0085] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for identifying crack width of asphalt pavement using radar amplitude characteristics, characterized by: The steps include: Divide the asphalt pavement into multiple detection lanes and determine the center line of each detection lane, collect radar data from each detection lane, analyze the longitude and latitude information of all radar data points, project the radar data points onto the satellite map, and obtain a complete radar projection map of the asphalt pavement; Based on the latitude and longitude coordinates of the centerline radar point Calculate the latitude and longitude coordinates of each radar point to be estimated : in, Angle is a straight line and the angle between due south, , is the average radius of the Earth, represents the magnitude of a vector, i.e. is the centerline radar point and radar points to be estimated The distance between Identify the edge of each crack in the radar projection image of the asphalt pavement, analyze the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack, and calculate the minimum distance from each crack edge point to the skeleton curve , after statistical sorting, the first average width of each crack is obtained Dp ; The complete radar projection image of the asphalt pavement is imported into the edge detection algorithm to identify the edge of each crack in the radar projection image of the asphalt pavement. Then, the detected crack edge is separated from the complete radar projection image of the asphalt pavement through a binarization operation to obtain the edge image of each crack and number them sequentially. Perform a thinning operation on the edge image of the crack to obtain a crack skeleton curve with a single pixel width, extract endpoints and inflection points on the crack skeleton curve and the edge image of the crack as feature points, use a feature point matching algorithm to match the feature points on the crack edge image with the feature points on the crack skeleton curve, and map the crack skeleton curve to the edge image of the crack; Traverse each crack edge point and all points on the skeleton curve, calculate the Euclidean distance between all points on the skeleton curve and a certain crack edge point, and obtain the minimum distance from each crack edge point to the skeleton curve , after statistical sorting, the first average width of each crack is obtained Dp , the first average width of the crack Dp For 2 ; in, c is the sequential numbering of the crack edge points of the same crack, The minimum distance from all crack edge points to the skeleton curve of the same crack The mean of Determine the boundary range of the crack, check the position of each radar point, mark it as the internal point of the crack or the data point without damage, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj , build a crack width analysis model; After identifying the edge of each crack in the radar projection image of the asphalt pavement, determine the boundary range of the crack, check the position of each radar point, and determine whether it is within the boundary range of the crack. If the radar point is within the boundary range of the crack, mark it as an internal point of the crack, and extract its coordinates and radar data to obtain the reflection amplitude of each internal point of each crack. The remaining radar points that are not marked as internal points of the cracks are marked as data points without defects. The reflection amplitudes of all data points without defects are extracted, and the mean reflection amplitudes of the data points without defects are obtained after statistics. ; Get the reflection amplitude of each internal point of each crack and the mean reflection amplitude of the data points without damage , calculate the average difference coefficient of crack amplitude Cj : in, i Indicates the sequential number of the points inside the crack, ; Get the average difference coefficient of crack amplitude of multiple cracks Cj and the first average width of cracks Dp , select part of the data as sample data, build a crack width analysis model based on a linear regression algorithm, and output the crack width analysis model after sample data training and testing; Check the position of each radar point, mark the crack skeleton point, and obtain the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point , the crack skeleton point amplitude difference coefficient Import the crack width analysis model to obtain the crack width at each crack skeleton point; Obtain a fused image of the crack skeleton curve mapped onto the edge image of the crack, check the position of each radar point to determine whether it is on the crack skeleton curve, if the radar point is on the crack skeleton curve, mark it as the skeleton point of the crack, extract its coordinates and radar data, and obtain the reflection amplitude of each crack skeleton point ; Get the reflection amplitude of each crack skeleton point and the mean reflection amplitude of the data points without damage , calculate the amplitude difference coefficient of the crack skeleton point : in, j Indicates the latitude and longitude position of the crack skeleton point; The amplitude difference coefficient of the crack skeleton point Import the crack width analysis model to obtain the crack width at each crack skeleton point.

Images