Method and system for automatically calculating expansion speed of test crack of asphalt mixture and estimating repair volume

Abstract

The invention discloses a method and a system for automatically calculating the expansion speed of an asphalt mixture test crack and estimating the repair volume. The method comprises the following steps of: preprocessing a test crack image by graying, smoothing and histogram equalization; a Canny edge detection algorithm, a BFMatcher feature point matching algorithm and a Lucas-Kanade feature point tracking algorithm are utilized to realize experimental crack tracking; the crack propagation speed is automatically calculated through time sequence analysis; acquiring three-dimensional point cloud data of the test crack and performing three-dimensional reconstruction; automatically calculating a crack orientation minimum bounding box; and estimating the repair volume of the crack. The method can effectively solve the problems of low detection efficiency and large subjectivity of the detection result caused by human eye observation or image recording in the crack propagation exploration process, improves the problem of inaccurate repair volume estimation after the test crack is generated, and improves the automation and refinement level of the test crack test.

Description

Technical Field

[0001] The invention relates to the technical field of material testing, and in particular to a method and a system for automatically calculating the growth rate of experimental cracks in asphalt mixtures and estimating the repair volume. Background Art

[0002] At present, asphalt mixture has been widely used in highway construction due to its good road performance and economy, and has become the mainstream choice in engineering implementation. In the process of asphalt mixture design and evaluation, experimental crack test is usually used to evaluate the crack resistance of the mixture. The introduction of experimental cracks helps to predict the problems faced by asphalt mixture during its service life, so as to guide the material selection and gradation design in road engineering, thereby improving road performance.

[0003] However, traditional mechanical analysis makes it difficult to characterize the expansion and fracture process and laws of experimental cracks. Today, the exploration of crack expansion process is usually achieved by human observation or image recording combined with mechanical testing. This type of detection technology not only has low detection efficiency, but also has highly subjective test results. In addition, the estimation of experimental crack volume based on manual measurement or image processing often only focuses on the crack length and ignores the crack depth, resulting in a large gap between the estimated value and the actual value. Therefore, there is an urgent need for automated and refined testing and research on the expansion, fracture and repair volume of experimental cracks in asphalt mixtures. Summary of the invention

[0004] In response to the above problems, the present invention provides a method and system for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixtures, which is based on image recognition technology and three-dimensional point cloud reconstruction technology and is improved in combination with traditional experimental crack mechanics tests. It is used to solve the problems of low detection efficiency and high subjectivity of detection results caused by human visual observation or image recording in the process of exploring crack growth, while improving the problem of inaccurate estimation of the repair volume after the occurrence of experimental cracks, strengthening the automation and refinement of experimental crack testing, and can be subsequently used for estimating the asphalt repair volume in asphalt pavement maintenance construction.

[0005] The method for automatically calculating the experimental crack growth rate of asphalt mixture and estimating the repair volume includes the following steps:

[0006] Asphalt mixture specimen pretreatment;

[0007] Collect orthophoto images of crack growth during asphalt mixture loading test;

[0008] Experimental crack image preprocessing;

[0009] Experimental crack tracking;

[0010] Automatic calculation of crack propagation speed;

[0011] Collect 3D point cloud data and 3D reconstruction of experimental cracks;

[0012] Automatic calculation of the minimum bounding box for crack orientation;

[0013] Crack repair volume estimation.

[0014] Optionally, the asphalt mixture specimen pretreatment comprises the following steps:

[0015] Spray white paint on the surface of asphalt mixture specimens;

[0016] Installing the asphalt mixture specimen on a testing machine, further, the testing machine can perform constant force loading;

[0017] The testing machine is used to adopt a loading method of constant force loading from the top of the asphalt mixture specimen, and the loading constant force is set.

[0018] Optionally, the collecting of orthophoto images of crack expansion during the asphalt mixture loading test comprises the following steps:

[0019] Install a high-definition camera to ensure that the field of view covers the asphalt mixture specimen and is located in front of the asphalt mixture specimen;

[0020] Setting up a lighting system to ensure that the ambient light of the asphalt mixture specimen is uniform and sufficient;

[0021] Adjusting the focal length of the high-definition camera to ensure that the asphalt mixture image is clear;

[0022] The testing machine is started to carry out a loading test and collect color images of the cracks.

[0023] Optionally, the experimental crack image preprocessing comprises the following steps:

[0024] Converting the crack color image into a crack grayscale image;

[0025] Using a Gaussian function weighted convolution kernel to smooth the crack grayscale image, eliminate noise, and highlight the characteristics of the crack;

[0026] Based on the histogram equalization principle, the grayscale values ​​of the crack grayscale image are redistributed to make them evenly distributed, thereby improving the global contrast of the image.

[0027] Optionally, the experimental crack tracking comprises the following steps:

[0028] Using Canny edge detection to identify and extract crack feature points of the preprocessed image to determine the initial position of the crack;

[0029] Using the BFMatcher feature point matching algorithm to identify and match the feature points in each subsequent frame of the image, and determine the crack position in the subsequent image;

[0030] The Lucas-Kanade optical flow method is used to track the crack feature points, calculate the image pixel motion, and track the changes of the cracks in consecutive frames;

[0031] The crack edge is further determined, and the position of the crack is updated by analyzing the feature points tracked by the optical flow method.

[0032] Optionally, the automatic calculation of the crack propagation speed comprises the following steps:

[0033] Crack length calculation, measuring the length of the crack tracking result;

[0034] Time series analysis, storing the data of the crack length changing with time;

[0035] The rate of change of crack length between adjacent time points is calculated, and the crack propagation speed is calculated.

[0036] Optionally, the collecting of crack three-dimensional point cloud data of the asphalt mixture specimen and the three-dimensional reconstruction include the following steps:

[0037] Using a three-dimensional laser scanner, the asphalt concrete specimen with cracks is scanned to obtain three-dimensional point cloud data;

[0038] Preprocessing the three-dimensional point cloud data to remove noise points, invalid points and outliers;

[0039] Using Canny edge detection, detecting crack edges on the projection image of the three-dimensional point cloud data;

[0040] Further, the crack points detected at the edge are extracted from the three-dimensional point cloud data to form a crack three-dimensional point cloud;

[0041] The three-dimensional point cloud of the crack is reconstructed into a three-dimensional mesh using a triangular mesh algorithm.

[0042] Optionally, the automatic calculation of the crack oriented minimum bounding box comprises the following steps:

[0043] Calculating the mean point of the three-dimensional crack point cloud data, further calculating the covariance matrix of the three-dimensional crack point cloud data, and further calculating the eigenvalues ​​and eigenvectors of the three-dimensional crack point cloud data;

[0044] Obtaining a new coordinate system according to the characteristic vectors of the three-dimensional crack point cloud data;

[0045] Projecting the three-dimensional crack point cloud data to the new coordinate system, and further acquiring new three-dimensional crack point cloud data in the new coordinate system;

[0046] The minimum and maximum coordinate values ​​of the new three-dimensional crack point cloud data are found, a minimum bounding box is further constructed, and the volume of the bounding box is calculated.

[0047] Optionally, the crack repair volume is estimated based on the bounding box volume.

[0048] Based on the same inventive concept, the present invention also proposes a system for automatically calculating crack propagation velocity and estimating repair volume, which comprises:

[0049] Image acquisition and preprocessing module, used to collect images of the experimental crack expansion process of asphalt mixture, and provide image data storage as well as grayscale processing, smoothing processing, and histogram equalization processing;

[0050] Crack recognition and tracking module, used to realize crack recognition and tracking, with built-in Canny edge detection algorithm, BFMatcher feature point matching algorithm, and Lucas-Kanade feature point tracking algorithm;

[0051] The time series analysis module is used to record the association between each stage of crack tracking and time, and store the data of crack length changes over time;

[0052] The crack extension speed automatic calculation module is used to automatically calculate the change rate of the crack length between adjacent time points, adjust the step size of adjacent time points, automatically draw the curve of the crack extension speed changing with time, and export the data of the measured crack extension speed;

[0053] Point cloud data acquisition and reconstruction module, used to collect 3D point cloud data of specific stages of experimental cracks in asphalt mixture, and provide point cloud data preprocessing and 3D point cloud reconstruction and storage of cracks;

[0054] The repair volume automatic estimation module is used to automatically estimate the repair volume, and has a built-in crack directional minimum bounding box automatic calculation algorithm.

[0055] The beneficial effects of the present invention are:

[0056] 1. The present invention utilizes a high-definition camera to quickly obtain image information of the dynamic expansion of experimental cracks in asphalt mixtures, and realizes rapid crack identification and positioning and contour morphology extraction through image processing algorithms, automatically obtains crack expansion time, and reduces human subjectivity in the test process.

[0057] 2. The present invention realizes experimental crack tracking through dynamic image recognition technology, and calculates the change rate of crack length between adjacent time points according to the time series analysis results, thereby automatically calculating the real-time crack expansion speed and improving the degree of automation of the detection process.

[0058] 3. The present invention combines a high-definition camera with a three-dimensional laser scanner to realize the positioning, contour extraction and three-dimensional reconstruction of experimental cracks in asphalt mixtures, and estimates the volume of cracks at each stage by adopting a bounding box algorithm, thereby providing a reference for the amount of asphalt repair in subsequent asphalt pavement maintenance construction, which is beneficial to saving pavement maintenance raw materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flow chart of a method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture provided for the implementation of the present invention;

[0060] Figure 2 A schematic diagram of experimental crack image preprocessing provided for the implementation of the present invention;

[0061] Figure 3 A schematic diagram of an experimental crack tracing process provided for the implementation of the present invention;

[0062] Figure 4 A schematic diagram of a system module for automatically calculating the experimental crack growth rate of asphalt mixture and estimating the repair volume is provided for the implementation of the present invention. DETAILED DESCRIPTION

[0063] The following is a further detailed description of a method and system for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to the present invention in conjunction with the accompanying drawings and specific embodiments:

[0064] Example:

[0065] like Figure 1 As shown, a method for automatically calculating the experimental crack growth rate of asphalt mixture and estimating the repair volume comprises the following steps:

[0066] S1: Asphalt mixture specimen pretreatment, specifically including the following steps:

[0067] S1.1: Spray white paint on the surface of asphalt mixture specimen;

[0068] S1.2: Install the asphalt mixture specimen described in S1.1 on a testing machine, which can perform constant force loading;

[0069] S1.3: Using the testing machine described in S1.2, adopt the loading method of constant force loading from the top of the asphalt mixture specimen described in S1.1, and set the loading constant force.

[0070] S2: Collecting orthophotos of crack propagation during the asphalt mixture loading test, specifically including the following steps:

[0071] S2.1: Install a high-definition camera to ensure that the field of view covers the asphalt mixture specimen described in S1.1 and is in front of it;

[0072] S2.2: Set up a lighting system to ensure that the ambient light of the asphalt mixture specimen described in S1.1 is uniform and sufficient;

[0073] S2.3: Adjust the focal length of the high-definition camera described in S2.1 to ensure that the image of the asphalt mixture described in S1.1 is clear;

[0074] S2.4: Start the testing machine described in S1.2 to perform a loading test and collect color images of the cracks.

[0075] S3: Experimental crack image preprocessing, which specifically includes the following steps:

[0076] S3.1: Convert the crack color image described in S2.4 into a crack grayscale image;

[0077] S3.2: Use a convolution kernel weighted by a Gaussian function to smooth the crack grayscale image described in S3.1, eliminate noise, and highlight the characteristics of the cracks. The convolution calculation formula is:

[0078]

[0079] Among them, I'(x,y) is the image pixel value after the convolution operation, k is the radius of the convolution kernel, and the convolution kernel size is (2k+1) 2 , G(i, j) is the weight of the Gaussian convolution kernel;

[0080] S3.3: Based on the principle of histogram equalization, the grayscale values ​​of the crack grayscale image described in S3.2 are redistributed to make them evenly distributed, such as Figure 2 As shown, the global contrast of the image is improved, and the histogram equalization calculation formula is:

[0081]

[0082] Among them, I e (x, y) is the image pixel value after histogram equalization, I'(x, y) is the image pixel value after convolution operation, CDF min is the non-zero minimum value of the cumulative distribution function, M×N is the total number of pixels in the image, and L is the number of gray levels.

[0083] S4: Experimental crack tracking, including the following steps:

[0084] S4.1: Use Canny edge detection to identify and extract crack feature points of the preprocessed image described in S3, and determine the initial position of the crack. The calculation formula for the edge gradient intensity and direction is:

[0085]

[0086] Among them, G x and G y are the gradients of the image in the x and y directions, G is the gradient strength, θ is the gradient direction, and the gradient strength of each pixel is compared with other points in the field, the local maximum value is retained, and the characteristic points of the crack edge are determined;

[0087] S4.2: Use the BFMatcher feature point matching algorithm to identify and match the feature points in each subsequent frame of the image described in S4.1 to determine the crack position in the subsequent image. The feature point matching calculation formula is:

[0088] D(p1,p2)=||desc(p1)-desc(p2)||

[0089] Among them, D(p1,p2) is the distance between the descriptors of feature points p1 and p2, desc(p) is the descriptor of feature point p. By calculating the distance between descriptors, the matching feature points in the image can be determined;

[0090] S4.3: Use the Lucas-Kanade optical flow method to track the crack feature points described in S4.2, calculate the image pixel movement, and track the changes of the cracks in consecutive frames. The crack tracking calculation formula is:

[0091]

[0092] In which, it is assumed that there is a pixel I(x, y, T) in the image at time T, which is at position I(x+u, y+v, T+t) at time T+t, (u, v) is the motion vector of each pixel, so as to track the changes of the crack in consecutive frames;

[0093] S4.4: Further determine the crack edge, such as Figure 3 As shown, the position of the crack is updated by analyzing the feature points tracked by the optical flow method described in S4.3.

[0094] S5: Automatic calculation of crack propagation speed, specifically including the following steps:

[0095] S5.1: Crack length calculation, measuring the length of the crack tracing results described in S4;

[0096] S5.2: Time series analysis, storing the data of the crack length change over time described in S5.1;

[0097] S5.3: Calculate the change rate of the crack length between adjacent time points and calculate the crack extension speed. The crack extension speed calculation formula is:

[0098]

[0099] Where v is the crack growth rate, L T1 and L T2 are the crack lengths at time T1 and T2, respectively.

[0100] S6: Collecting experimental crack 3D point cloud data and 3D reconstruction, specifically including the following steps:

[0101] S6.1: Scanning the cracked asphalt concrete specimen using a three-dimensional laser scanner to obtain three-dimensional point cloud data;

[0102] S6.2: preprocessing the three-dimensional point cloud data to remove noise points, invalid points and outlier points;

[0103] S6.3: Using Canny edge detection, detect crack edges on the projected image of the three-dimensional point cloud data described in S6.2;

[0104] S6.4: extracting crack points detected at the edge from the three-dimensional point cloud data to form a crack three-dimensional point cloud;

[0105] S6.5: Reconstruct the three-dimensional point cloud of the crack described in S6.4 into a three-dimensional mesh using a triangular mesh algorithm.

[0106] S7: Automatic calculation of the crack oriented minimum bounding box, specifically including the following steps:

[0107] S7.1: Calculate the mean point of the three-dimensional crack point cloud data described in S6, further calculate the covariance matrix of the three-dimensional crack point cloud data described in S6, and further calculate the eigenvalues ​​and eigenvectors of the three-dimensional crack point cloud data described in S6;

[0108] S7.2: Obtain a new coordinate system based on the feature vectors of the three-dimensional crack point cloud data described in S7.1;

[0109] S7.3: Projecting the three-dimensional crack point cloud data described in S6 to the new coordinate system described in S7.2, and further obtaining new three-dimensional crack point cloud data in the new coordinate system;

[0110] S7.4: Find the minimum and maximum coordinate values ​​of the new three-dimensional crack point cloud data described in S7.3, and further construct a minimum bounding box. The bounding box volume calculation formula is:

[0111] V=(max'x-min'x)×(max'y-min'y)×(max'z-min'z)

[0112] Among them, max'x and min'x are the maximum and minimum coordinates in the x direction, max'y and min'y are the maximum and minimum coordinates in the y direction, and max'z and min'z are the maximum and minimum coordinates in the z direction.

[0113] S8: The crack repair volume is calculated according to the bounding box volume described in S7.

[0114] Based on the same inventive concept, the present invention also proposes a system for automatically calculating crack propagation velocity and estimating repair volume, such as Figure 4 As shown, it includes:

[0115] Image acquisition and preprocessing module, used to collect images of the experimental crack expansion process of asphalt mixture, and provide image data storage as well as grayscale processing, smoothing processing, and histogram equalization processing;

[0116] Crack recognition and tracking module, used to realize crack recognition and tracking, with built-in Canny edge detection algorithm, BFMatcher feature point matching algorithm, and Lucas-Kanade feature point tracking algorithm;

[0117] The time series analysis module is used to record the association between each stage of crack tracking and time, and store the data of crack length changes over time;

[0118] The crack extension speed automatic calculation module is used to automatically calculate the change rate of the crack length between adjacent time points, adjust the step size of adjacent time points, automatically draw the curve of the crack extension speed changing with time, and export the data of the measured crack extension speed;

[0119] Point cloud data acquisition and reconstruction module, used to collect 3D point cloud data of specific stages of experimental cracks in asphalt mixture, and provide point cloud data preprocessing and 3D point cloud reconstruction and storage of cracks;

[0120] The repair volume automatic estimation module is used to automatically estimate the repair volume, and has a built-in crack directional minimum bounding box automatic calculation algorithm.

[0121] Based on the same inventive concept, an embodiment of the present invention provides an electronic device (computer, server, smart phone, etc.), which includes a memory and a processor, the memory stores a computer program, the computer program is configured to be executed by the processor, and the computer program includes instructions for executing each step in the method of the present invention.

[0122] The present invention does not elaborate on some of the well-known technologies that belong to those skilled in the art.

[0123] The examples of the present invention are described in detail above in conjunction with the embodiments, but the present invention is not limited to the above examples. Various changes can be made within the knowledge scope of ordinary technicians in the field without departing from the purpose of the present invention, and should also be regarded as the protection scope of the present invention.

Claims

1. A method for automatically calculating the experimental crack growth rate of asphalt mixture and estimating the repair volume, characterized in that: The following steps are involved: Asphalt mixture specimen pretreatment; Collect orthophoto images of crack growth during asphalt mixture loading test; Experimental crack image preprocessing; Experimental crack tracking; Automatic calculation of crack propagation speed; Collect 3D point cloud data and 3D reconstruction of experimental cracks; Automatic calculation of the minimum bounding box for crack orientation; Crack repair volume estimation.

2. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 1 is characterized in that: The asphalt mixture specimen pretreatment comprises the following steps: Spray white paint on the surface of asphalt mixture specimens; Install the asphalt mixture specimen on the testing machine; Using the testing machine, a loading method of constant force loading from the top of the asphalt mixture specimen is adopted; The method of collecting the orthophoto of crack expansion during the asphalt mixture loading test comprises the following steps: Install a high-definition camera to ensure that the field of view covers the asphalt mixture specimen and is located in front of the asphalt mixture specimen; Setting up a lighting system to ensure that the ambient light of the asphalt mixture specimen is uniform and sufficient; Adjusting the focal length of the high-definition camera to ensure a clear image of the asphalt mixture; Starting the testing machine to perform a loading test and simultaneously collecting a color image of the crack; The experimental crack image preprocessing comprises the following steps: Converting the crack color image into a crack grayscale image; Using a Gaussian function weighted convolution kernel to smooth the crack grayscale image, eliminate noise, and highlight the characteristics of the crack; Based on the histogram equalization principle, the grayscale values ​​of the crack grayscale image are redistributed to make them evenly distributed, thereby improving the global contrast of the image.

3. The experimental crack image preprocessing method according to claim 2, characterized in that: The convolution kernel weighted by the Gaussian function is used to smooth the crack grayscale image, and the convolution calculation formula is: Among them, I'(x,y) is the image pixel value after the convolution operation, k is the radius of the convolution kernel, and the convolution kernel size is (2k+1) 2 , G(i, j) is the weight of the Gaussian convolution kernel; The grayscale value of the crack grayscale image is redistributed based on the histogram equalization principle. The histogram equalization calculation formula is: Among them, I e (x, y) is the image pixel value after histogram equalization, I'(x, y) is the image pixel value after convolution operation, CDF min is the non-zero minimum value of the cumulative distribution function, M×N is the total number of pixels in the image, and L is the number of gray levels.

4. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 1 is characterized in that: The experimental crack tracking comprises the following steps: Using Canny edge detection to identify and extract crack feature points of the preprocessed image to determine the initial position of the crack; Using the BFMatcher feature point matching algorithm to identify and match the feature points in each subsequent frame of the image, and determine the crack position in the subsequent image; The Lucas-Kanade optical flow method is used to track the crack feature points, calculate the image pixel motion, and track the changes of the cracks in consecutive frames; The crack edge is further determined, and the position of the crack is updated by analyzing the feature points tracked by the optical flow method.

5. The experimental crack tracing method according to claim 4, characterized in that: The calculation formula for the edge gradient strength and direction identified by the Canny edge detection is: Among them, G x and G y are the gradients of the image in the x and y directions, G is the gradient strength, θ is the gradient direction, and the gradient strength of each pixel is compared with other points in the field, the local maximum value is retained, and the characteristic points of the crack edge are determined; The BFMatcher feature point matching calculation formula is: D(p1,p2)=||desc(p1)-desc(p2)|| Among them, D(p1,p2) is the distance between the descriptors of feature points p1 and p2, desc(p) is the descriptor of feature point p. By calculating the distance between descriptors, the matching feature points in the image can be determined; The Lucas-Kanade optical flow method crack tracking calculation formula is: In it, it is assumed that there is a pixel point I(x, y, T) in the image at time T, and it is at position I(x+u, y+v, T+t) at time T+t, (u, v) is the motion vector of each pixel, thereby tracking the changes of the crack in consecutive frames.

6. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 1 is characterized in that: The automatic calculation of the crack propagation speed comprises the following steps: Crack length calculation, measuring the length of the crack tracking result; Time series analysis, storing the data of the crack length changing with time; Calculate the change rate of crack length between adjacent time points and the crack expansion speed; The calculation formula for calculating the change rate of the crack length between adjacent time points and the crack extension speed is: Where v is the crack growth rate, L T1 and L T2 are the crack lengths at time T1 and T2, respectively.

7. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 1 is characterized in that: The acquisition of experimental crack 3D point cloud data and 3D reconstruction comprises the following steps: Using a three-dimensional laser scanner, the asphalt concrete specimen with cracks is scanned to obtain three-dimensional point cloud data; Preprocessing the three-dimensional point cloud data to remove noise points, invalid points and outliers; Using Canny edge detection, detecting crack edges on the projection image of the three-dimensional point cloud data; Extracting crack points detected at the edge from the three-dimensional point cloud data to form a crack three-dimensional point cloud; The three-dimensional point cloud of the crack is reconstructed into a three-dimensional mesh using a triangular mesh algorithm.

8. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 1 is characterized in that: The automatic calculation of the crack oriented minimum bounding box comprises the following steps: Calculating the mean point of the three-dimensional crack point cloud data, further calculating the covariance matrix of the three-dimensional crack point cloud data, and further calculating the eigenvalues ​​and eigenvectors of the three-dimensional crack point cloud data; Obtaining a new coordinate system according to the characteristic vectors of the three-dimensional crack point cloud data; Projecting the three-dimensional crack point cloud data to the new coordinate system, and further acquiring new three-dimensional crack point cloud data in the new coordinate system; Find the minimum and maximum coordinate values ​​of the new three-dimensional crack point cloud data, further construct a minimum bounding box, and calculate the volume of the bounding box; The formula for calculating the volume of the minimum bounding box is as follows: V=(max'x-min'x)×(max'y-min'y)×(max'z-min'z) Among them, max'x and min'x are the maximum and minimum coordinates in the x direction, max'y and min'y are the maximum and minimum coordinates in the y direction, and max'z and min'z are the maximum and minimum coordinates in the z direction.

9. The method for automatically calculating the experimental crack growth rate and estimating the repair volume of asphalt mixture according to claim 8, characterized in that: The seam repair volume is estimated based on the bounding box volume.

10. An automatic calculation system for experimental crack growth rate and repair volume estimation of asphalt mixture, characterized in that: include: Image acquisition and preprocessing module: used to collect images of the experimental crack expansion process of asphalt mixture, and provide image data storage as well as grayscale processing, smoothing processing, and histogram equalization processing; Crack identification and tracking module: used to realize crack identification and tracking, with built-in Canny edge detection algorithm, BFMatcher feature point matching algorithm, and Lucas-Kanade feature point tracking algorithm; Time series analysis module: used to record the relationship between each stage of crack tracking and time, and store the data of crack length changes over time; Automatic calculation module for crack extension speed: used to automatically calculate the change rate of crack length between adjacent time points, adjust the step length of adjacent time points, automatically draw the curve of crack extension speed changing with time, and export the data of measured crack extension speed; Point cloud data acquisition and reconstruction module: used to collect 3D point cloud data of specific stages of experimental cracks in asphalt mixtures, and provide point cloud data preprocessing and 3D point cloud reconstruction and storage of cracks; Automatic repair volume estimation module: used to automatically estimate the repair volume, with a built-in crack directional minimum bounding box automatic calculation algorithm.

Images