Pavement rut depth measuring method

Abstract

The invention belongs to the technical field of pavement rut depth measurement, and discloses a pavement rut depth measurement method. Obtaining a size scaling coefficient and an installation deviation angle through experimental calibration; acquiring an image, cutting the image, preprocessing the image, extracting a light strip center line, and converting coordinates to obtain a cross section profile curve; calculating an envelope line according to the point-by-point coordinates of the cross section profile curve; respectively calculating the left and right sides of the cross section, and combining the calibrated installation deviation angle to obtain the rut depth. The method provided by the invention can effectively solve the problem of accuracy loss of the rut depth calculation result caused by angle errors in the installation process. Meanwhile, the rut depth of the left wheel track and the rut depth of the right wheel track can be calculated at the same time, manual operation is simplified, and the requirement of the current industry standard is met. The method is easy to implement and higher in precision, and the application requirement is met.

Description

Technical Field

[0001] The present invention relates to the technical field of pavement rut depth measurement, and particularly to a method for measuring pavement rut depth. Background Art

[0002] After years of rapid development, the scale of China's highway transportation network has been continuously expanding. By the end of 2023, the total highway mileage reached 5.4368 million kilometers. However, most of the roads built in the early stage have entered the middle and late stages of their life cycles, and the maintenance pressure on the roads has been increasing year by year. The highway road network has generally entered the maintenance stage, and China's road network is gradually shifting from large-scale new road paving to large-scale old road maintenance. Research and engineering practice have shown that from the perspective of the whole life cycle, the proportion of highway maintenance investment in all highway investments will far exceed the investment in the design and construction period. Therefore, how to carry out efficient, economical, and timely maintenance management of highway pavements so as to keep the roads in good condition is the top priority for improving the service life of roads, maintaining traffic efficiency, and saving maintenance funds.

[0003] Ruts are the wheel imprints left by vehicles after driving on the road surface. The following multiple devices can be used for automatic detection of pavement ruts. (1) Beam-type multi-sensors: Measure the relative elevations at different positions of the cross-section through multiple distance sensors on the same reference plane, not less than 13. (2) Scanning laser sensors: High-speed scan the cross-section of the road surface through one or more laser emitters to obtain a continuous cross-sectional curve. (3) Optical images: Use a line laser emitter and a high-speed image acquisition device to obtain data, and determine the rut depth by analyzing the deformation of light.

[0004] The solution described in the Chinese patent "CN117888429A A pavement rut anomaly detection system and method based on line structured light" is a pavement rut anomaly detection system and method based on line structured light, including a longitudinal distance measurement sensor, a two-dimensional linear array camera, and a line structured light generator. It includes methods such as obtaining pavement rut data images, screening out abnormal ruts from the pavement rut data images according to a preset threshold, performing data verification on the abnormal ruts to obtain a data verification structure, and determining the evaluation model to which the pavement rut data image belongs according to the data verification result. It does not introduce the calibration of rut depth measurement equipment and the numerical calculation method.

[0005] The solution described in the Chinese patent "CN119129407A A method and device, system, storage medium for predicting road rut depth" mainly trains the rut data of the road through a neural network model optimized by a genetic algorithm, so as to achieve the purpose of predicting the rut depth through the rut depth data, which is different from the solution in this scheme that automatically calculates the rut depths of the left wheel track belt and the right wheel track belt by using the envelope line method. Summary of the Invention

[0006] The object of the present invention is to provide a method for measuring the rut depth of a road surface, which is divided into two stages: calibration of the rut depth measurement device and calculation of the rut depth value. This method can reduce the numerical calculation error caused by the error existing in the installation process of the measurement device, and can also solve the automatic calculation requirement for the rut depth of the left and right wheel tracks, meeting the relevant requirements of the industry standard for rut depth measurement.

[0007] The technical solution of the present invention is as follows: A method for measuring the rut depth of a road surface, installing a rut depth measurement device for the road surface, and obtaining a size scaling coefficient and an installation deviation angle through experimental calibration;

[0008] Collecting images; adjusting the exposure mode of the industrial camera to "auto exposure", and realizing camera photographing and image storage through a hard trigger method;

[0009] Image cropping; cropping the image according to the result of the "region of interest range";

[0010] Image preprocessing; performing binarization and morphological processing on the cropped image;

[0011] Extracting the center line of the light strip; traversing column by column, extracting the light strip range, and calculating the row number where the center line of the light strip is located;

[0012] Coordinate transformation; performing coordinate transformation according to the center pixel value of the light strip, the internal parameters of the camera, the "size scaling coefficient" obtained by calibration and the "laser triangulation" formula to obtain the cross-sectional profile curve of the road;

[0013] Calculating the envelope line; calculating the envelope line according to the point-by-point coordinates of the cross-sectional profile curve of the road;

[0014] Calculating the rut depth, calculating respectively on the left and right of the road cross-section according to the set center threshold, and obtaining the rut depth in combination with the calibrated installation deviation angle.

[0015] The rut depth measurement device for the road surface includes a red light laser, a red light filter, a near-infrared enhanced industrial camera, and a supporting cable;

[0016] The red light filter is installed on the surface of the lens of the near-infrared enhanced industrial camera. CAM represents the near-infrared enhanced industrial camera, LAS is the red light laser, A is the intersection point of the optical axis of the red light laser and the near-infrared enhanced industrial camera. The horizontal plane where the intersection line of the optical axis of the near-infrared enhanced industrial camera and the red light laser is located is used as the reference plane. O is the position of the lens of the near-infrared enhanced industrial camera, H is the height of the lens of the near-infrared enhanced industrial camera from the reference plane, β is the angle between the near-infrared enhanced industrial camera and the reference plane, and α is the angle between the red light laser and the reference plane;

[0017] For a point C on the ground, draw a perpendicular line through the point to the reference plane, with the foot of the perpendicular being B, and h being the distance between the point and the reference plane; x is the offset size of point C in the image plane, f is the focal length of the camera, k is the conversion factor, and p is the offset pixel of point C in the image plane;

[0018] The relationship between ground deformation and camera imaging pixel position change is obtained through the following formula:

[0019]

[0020] The size scaling factor is calculated as follows:

[0021] Step A. extracting the rutting plate area of the rutting plate image;

[0022] Step B. Image cropping: The cropped image ensures that the wheel track plate is within the minimum bounding rectangle of the cropped image area;

[0023] Step C. measuring the pixels corresponding to the highest area and the pixels corresponding to the lowest area of the rutting plate based on the cropped image;

[0024] Step D. Substitution Perform calculations;

[0025] Step E. Calculate the distance between the point obtained in step D and the reference line and the true depth h of the rutting plate real The quotient between them is used to calculate the size scaling factor k′;

[0026]

[0027] h1 represents the distance between the point calculated in the i-th experimental step D and the reference line surface, where i is an integer from 1 to n.

[0028] The calculation method of the calibrated installation deviation angle is:

[0029] Step A. extracting the laser line area in the static image;

[0030] Step B. measuring the row pixels and column pixels at the leftmost and rightmost ends of the laser line area;

[0031] Step C. Calculate the horizontal coordinate and vertical coordinate of the leftmost pixel point and the rightmost pixel point of the laser line; the horizontal coordinate is obtained by measurement, and the vertical coordinate is calculated using the laser triangulation formula;

[0032] Step D. Calculate the calibrated installation deviation angle using an angle calculation formula;

[0033]

[0034] Among them, angle represents the calibrated installation deviation angle; x represents the abscissa of the pixel point, and y represents the ordinate of the pixel point. The left and right points are distinguished by different subscripts of left and right.

[0035] The static image is an image obtained by photographing the ground with a horizontal shot and no obvious undulating area; the rut board image is obtained by photographing with a near-infrared enhanced industrial camera. During the photographing process, it is required that the laser line must pass through the horizontal areas at both ends and the sunken area in the middle of the rut board to ensure the correctness of the reference values, and several rut board images are collected repeatedly to avoid the influence of random errors; the rut board during the calibration process is an experimental material made during the experiment, with horizontal ends and a concave middle, used to simulate the actual rut shape.

[0036] The calculation method of the range of the region of interest is as follows: According to the 80mm standard measurement of the rut board, calculate the upper and lower boundary ranges of the red laser line in the imaging region, and use this boundary range as the range of the region of interest for subsequent image cropping.

[0037] The calculation of the rut depth is specifically as follows:

[0038] 1) Traverse the endpoints of the envelope line point by point, and set the first point as the starting point start of the segment p , and set the next point as the ending point end of the segment p ;

[0039] 2) Calculate the distance distance between the starting point and the ending point of the segment envelope ;

[0040]

[0041] 3) Calculate the angle angle between the line segment formed by the starting point and the ending point of the segment and the horizontal direction envelope ;

[0042] 4) Traverse each point between the starting point and the ending point of the segment on the center line of the light strip, and perform the following calculations:

[0043] 4.1) Calculate the cross product result of this point md p and the vector formed by the starting point and the ending point of the segment. The area of the parallelogram formed by these two vectors is as follows:

[0044]

[0045] 4.2) Calculate the vertical distance distance from the point md p to the envelope line vertical ;

[0046] distance vertical = cross / distanceenvelope

[0047] 4.2) Calculate the point md p The vertical distance distance from the envelope upright ; Considering the calibrated installation error angle, ensure that the rut depth solving direction is perpendicular to the road;

[0048] distance upright = distance vertical / cos(angle envelope - angle)

[0049] 4.3) Calculate the vertical distance from each point on the center line of the light strip between the starting point and the ending point of the segment to the envelope point by point until the maximum distance is found, which is used as the rut depth of this segment.

[0050] Advantages of the present invention: The present invention proposes an equipment installation deviation angle and a scale scaling coefficient. By calibrating the angle deviation existing in the equipment installation process and the scale scaling coefficient involved in the formula solving, the installation deviation angle and the dimension scaling coefficient can be simply solved, improving the measurement accuracy of the rut depth. At the same time, through the segmented solving algorithm based on the envelope line and combined with the equipment installation error angle parameter, the rut depths of the left and right wheel tracks can be calculated simultaneously, better meeting the requirements of industry standards and application needs.

[0051] For the data acquisition equipment that requires separate installation of cameras and laser lines, the method of the present invention can effectively solve the problem of accuracy loss of the rut depth calculation result caused by the angle error existing in the installation process. At the same time, the present invention can calculate the rut depths of the left and right wheel tracks simultaneously, simplifying the manual operation and meeting the requirements of current industry standards. The present invention is simple to implement and has higher accuracy, meeting the application requirements. Brief Description of the Drawings

[0052] Figure 1 It is a schematic diagram of the installation of the pavement rut depth measurement equipment;<able>

[0053] Figure 2 It is a flowchart of the calibration process;

[0054] Figure 3 It is a flowchart of the rut depth numerical calculation.

[0055] Figure 4 It is a schematic diagram of a static image.

[0056] Figure 5 It is a schematic diagram of an image containing a rut board.

[0057] Figure 6 It is the effect of image binarization.

[0058] Figure 7 The effect after morphological processing, with the kernel size selected as (5, 5).

[0059] Figure 8 The effect of light stripe center extraction.

[0060] Figure 9 The effect after coordinate transformation.

[0061] Figure 10 The architecture diagram of the rut measurement method. Specific implementation mode

[0062] A method for measuring the rut depth of a road surface, installing a road surface rut depth measurement device, and obtaining a size scaling coefficient and an installation deviation angle through experimental calibration;

[0063] Collect images; adjust the exposure mode of the industrial camera to "auto exposure", and realize camera photographing and image storage through the hard trigger mode;

[0064] Image cropping; crop the image according to the result of the "region of interest range";

[0065] Image preprocessing; perform binarization and morphological processing on the cropped image;

[0066] Extract the center line of the light stripe; traverse column by column, extract the light stripe range, and calculate the row number where the center line of the light stripe is located;

[0067] Coordinate transformation; perform coordinate transformation according to the light stripe center pixel value, the camera internal parameters, the "size scaling coefficient" obtained by calibration, and the "laser triangulation" formula to obtain the cross-sectional profile curve;

[0068] Envelope calculation; calculate the envelope according to the point coordinates of the cross-sectional profile curve;

[0069] Rut depth calculation, calculate separately on the left and right of the cross section according to the set center threshold, and combine the calibrated installation deviation angle to obtain the rut depth.

[0070] Furthermore, the road surface rut depth measurement method mentioned in the present invention adopts the "optical image" method, and the basic principle is "laser triangulation". The measurement device consists of a red light laser line, a near-infrared enhanced industrial camera, a lens, and a supporting cable.

[0071] Figure 1 Shown is the installation schematic diagram of the road surface rut depth measurement device of the present invention.

[0072] The CAM represents a near-infrared enhanced industrial camera, the LAS is a red light laser, A is the intersection point of the line laser and the optical axis of the near-infrared enhanced industrial camera, the horizontal plane where the intersection line of the optical axis of the near-infrared enhanced industrial camera and the line laser is located is the reference plane, O is the position of the lens of the near-infrared enhanced industrial camera, H is the height of the lens of the near-infrared enhanced industrial camera from the reference plane, β is the angle between the near-infrared enhanced industrial camera and the reference plane, and α is the angle between the line laser and the reference plane.

[0073] For a certain point C on the ground, draw a perpendicular line from this point to the reference line, and the foot of the perpendicular is B. h is the distance between this point and the reference plane. x is the offset dimension of point C in the image plane, f is the focal length of the camera, k is the conversion coefficient, and p is the offset pixel of point C in the image plane.

[0074] Through the following formula, the relationship between the ground deformation and the change in the pixel position of the camera imaging can be obtained.

[0075]

[0076] Figure 2 It is the calibration flow chart of the rut depth measurement device in the technical solution of the present invention. Through this part of the calibration process, the calibrated size scaling coefficient can be determined. At the same time, the problem that the line laser imaging is not horizontal due to the installation angle error can be solved.

[0077] 1) Step one: Collect images. This part includes collecting static images and rut board images.

[0078] The static image refers to the image obtained by shooting the ground in a horizontal and non-significantly undulating area; as Figure 4 shown,

[0079] The rut board refers to the experimental material made during the experiment, with both ends horizontal and the middle concave, used to simulate the actual rut shape.

[0080] During the process of shooting the rut board image, it is required that the laser line must pass through the horizontal areas at both ends and the sunken area in the middle of the rut board to ensure the correctness of the reference values, and 10 images are collected repeatedly to avoid the influence of random errors; as Figure 5 shown;

[0081] 2) Step two: Obtain the coefficients. This part includes calculating the size scaling coefficient (based on the rut board image) and the installation deviation angle (based on the static image);

[0082] 2.1) The calculation method of the size scaling coefficient is as follows: A. Extract the rut board area of the rut board image. B. Image cropping. C. Measure the corresponding pixels of the highest (horizontal) area and the lowest (bottom of the groove) area of the rut board in the cropped image. D. Substitute into Perform calculations. E. Calculate the quotient between the distance between the point obtained in step D and the reference line plane and the true depth of the rut board, and calculate the dimension scaling coefficient k′; as Figure 6 shown;

[0083]

[0084] 2.2) The calculation method for the installation deviation angle is as follows: A. Extract the laser line area in the static image. B. Measure the row pixels and column pixels at the leftmost and rightmost ends of the laser line area. C. Calculate the abscissa and ordinate of the leftmost pixel point of the laser line and the abscissa and ordinate of the rightmost pixel point. The abscissa is obtained by measurement, and the ordinate is calculated by the laser triangulation formula. D. Perform calculations through the angle calculation formula.

[0085]

[0086] Among them, angle represents the installation deviation angle. x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, and the left and right points are distinguished by different subscripts of left and right.

[0087] The calculation method for the range of the region of interest is as follows: Calculate the upper and lower boundary ranges of the laser line in the imaging area according to the 80mm standard measurement of the rut board, and use this range as the region of interest range for subsequent image cropping.

[0088] Figure 3 is the flowchart for calculating the rut depth value in the technical solution of the present invention; as Figure 6 — Figure 9 shown;

[0089] Step 1: Collect images. Adjust the exposure mode of the industrial camera to "auto exposure", and achieve camera shooting and image storage through the hard trigger method.

[0090] Step 2: Image cropping. Crop the image according to the obtained "region of interest range" result.

[0091] Step 3: Image preprocessing. Perform binaryzation and morphological processing on the image.

[0092] Step 4: Extract the center line of the light strip. Traverse column by column, extract the light strip range, and calculate the row number where the center line is located.

[0093] Step 5: Coordinate transformation. Perform coordinate transformation according to the center pixel value of the light strip, the camera internal parameters, the "dimension scaling coefficient" obtained by calibration, and the "laser triangulation" formula to obtain the cross-sectional profile curve.

[0094] Step 6: Calculate the envelope line. Calculate the envelope line according to the point-by-point coordinates of the cross-sectional profile curve.

[0095] Step 7: Rut depth calculation. According to the set center threshold, the following steps are calculated separately for the left and right of the cross-section:

[0096] 7.1) Traverse the envelope endpoints point by point. Set the first point as the start point of the segment (start p ), and the next point as the end point of the segment (end p ).

[0097] 7.2) Calculate the distance between the start point and the end point of the segment (distance envelope );

[0098] 7.3) Calculate the angle of the line segment formed by the start point and the end point of the segment (angle envelope );

[0099] 7.4) Traverse each point of the light strip center line between the start point and the end point of the segment, and perform the following calculations:

[0100] 7.5) Calculate the cross product result of the vector formed by this point (md p ) and the vector formed by the start point and the end point of the segment, which is the area of the parallelogram formed by these two vectors. The formula is as follows:

[0101]

[0102] 7.6) Calculate the perpendicular distance (distance p ) from the point md vertical to the envelope line.

[0103] distance vertical = cross / distance envelope

[0104] 7.7) Calculate the vertical distance (distance p ) from the point md upright to the envelope line. This part needs to consider the installation error angle obtained by calibration to ensure that the rut depth solving direction is perpendicular to the road.

[0105] distance upright = distance vertical / cos(angle envelope - angle)

[0106] 7.8) Calculate the vertical distance from each point of the light strip center line between the start point and the end point of the segment to the envelope line point by point until the maximum distance is found, which is used as the rut depth of this segment.

Claims

1. A method for measuring the rut depth of a road surface, characterized in that, Install a pavement rut depth measurement device, and obtain the size scaling coefficient and installation deviation angle through experimental calibration; Collect images; adjust the exposure mode of the industrial camera to "auto exposure", and achieve camera shooting and image storage through hard triggering; Image cropping; crop the image according to the result of the "region of interest range"; Image preprocessing; Perform binaryzation and morphological processing on the cropped image; Extract the center line of the light strip; traverse column by column, extract the range of the light strip, and calculate the row number where the center line of the light strip is located; Coordinate transformation; perform coordinate transformation according to the center pixel value of the light strip, the camera internal parameters, the "size scaling coefficient" obtained by calibration, and the "laser triangulation" formula to obtain the road cross-section profile curve; Envelope calculation; calculate the envelope according to the point coordinates of the road cross-section profile curve; Rut depth calculation, calculate the left and right sides of the road cross-section respectively according to the set center threshold, and combine the calibrated installation deviation angle to obtain the rut depth.

2. The method for measuring the rut depth of a road surface according to claim 1, wherein The pavement rut depth measurement device includes a red light laser, a red light filter, a near-infrared enhanced industrial camera, and a supporting cable; The red light filter is installed on the surface of the lens of the near-infrared enhanced industrial camera. CAM represents the near-infrared enhanced industrial camera, LAS is the red light laser, A is the intersection point of the optical axis of the red light laser and the near-infrared enhanced industrial camera. The horizontal plane where the intersection line of the optical axis of the near-infrared enhanced industrial camera and the red light laser is the reference plane, O is the position of the lens of the near-infrared enhanced industrial camera, H is the height of the lens of the near-infrared enhanced industrial camera from the reference plane, β is the angle between the near-infrared enhanced industrial camera and the reference plane, and α is the angle between the red light laser and the reference plane; For a certain point C on the ground, draw a perpendicular line from this point to the reference plane, and the foot of the perpendicular is B. h is the distance between this point and the reference plane; x is the offset dimension of point C in the image plane, f is the camera focal length, k is the conversion coefficient, and p is the offset pixel of point C in the image plane; Through the following formula, obtain the relationship between ground deformation and the change of camera imaging pixel position; 3. The pavement rut depth measurement method according to claim 2, characterized in that, The calculation method of the size scaling coefficient is: Step A. Extract the rut board area of the rut board image; Step B. Image cropping; ensure that the rut board is within the minimum bounding rectangle of the cropped image area after cropping; Step C. Measure the pixels corresponding to the highest area and the lowest area of the rut board based on the cropped image; Step D. Substitute Perform the calculation; Step E. Calculate the quotient of the distance between the point obtained in Step D and the reference line plane and the true depth h of the rut board, and calculate the dimension scaling factor k′; real ​ h1 represents the distance between the point calculated in step D of the i-th experiment and the reference line plane, where the value of i is an integer from 1 to n.

4. The pavement rut depth measurement method according to claim 2, characterized in that The calculation method of the calibrated installation deviation angle is: Step A. Extract the laser line area in the static image; Step B. Measure the row pixels and column pixels at the leftmost and rightmost ends of the laser line area; Step C. Calculate the abscissa and ordinate of the leftmost pixel point of the laser line and the abscissa and ordinate of the rightmost pixel point; the abscissa is obtained by measurement, and the ordinate is calculated by the laser triangulation formula; Step D. Calculate the calibrated installation deviation angle through the angle calculation formula; Among them, angle represents the calibrated installation deviation angle; x represents the abscissa of the pixel point, y represents the ordinate of the pixel point, and the left and right points are distinguished by different subscripts left and right.

5. The pavement rut depth measurement method according to claim 4, characterized in that, The static image is an image obtained by photographing the ground with a horizontal view and no obvious undulating areas; the rut board image is obtained by shooting with a near-infrared enhanced industrial camera. During the shooting process, it is required that the laser line must pass through the horizontal areas at both ends and the sunken area in the middle of the rut board to ensure the correctness of the reference values, and several rut board images are repeatedly collected to avoid the influence of random errors; the rut board in the calibration process is an experimental material made in the experimental process, with horizontal ends and a concave middle, used to simulate the actual rut shape.

6. The method for measuring the rut depth of a road surface according to claim 1, wherein The calculation method of the range of the region of interest is as follows: Measure the rut board according to the 80mm standard, calculate the upper and lower boundary ranges of the red laser line in the imaging area, and use this boundary range as the range of the region of interest for subsequent image cropping.

7. The method for measuring the rut depth of a road surface according to claim 1, characterized in that The calculation of the rut depth is specifically as follows: 1) Traverse the envelope endpoints point by point, set the first point as the start point start of the segment p , and set the next point as the end point end of the segment p ; 2) Calculate the distance distance between the starting point and the ending point of the segment envelope ; 3) Calculate the angle angle between the line segment formed by the starting point and the ending point of the segment and the horizontal direction envelope ; 4) Traverse each point between the starting point and the ending point of the light strip center line segment, and perform the following calculations: 4.1) Calculate the point md p The cross product result of the vector formed by the starting point and the ending point of the segment, which is the area of the parallelogram formed by these two vectors. The formula is as follows: 4.2) Calculate the point md p The perpendicular distance distance from the envelope vertical ; distance vertical = cross / distance envelope 4.2) Calculate the point md p The vertical distance distance from the envelope upright ; Considering the calibrated installation error angle, ensure that the rut depth solving direction is perpendicular to the road; distance upright = distance vertical / cos(angle envelope - angle) 4.3) Calculate the vertical distance from each point between the starting point and the ending point of the light strip center line segment to the envelope line point by point until the maximum distance is found, which is used as the rut depth of this segment.

Images