Super high-speed three-dimensional laser scanning pavement anti-skid performance evaluation method and system

Abstract

The application discloses a kind of super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method and system, the method is scanned to target pavement using three-dimensional laser scanner, obtains pavement three-dimensional point cloud original data, pre-processes pavement three-dimensional point cloud original data, constructs pavement three-dimensional texture model according to the pavement three-dimensional point cloud data after pre-processing, by constructing the datum plane of target pavement, and the datum plane is projected to the pavement three-dimensional texture model, based on the pavement three-dimensional texture model after projection, extract a plurality of characteristic parameters related to pavement anti-skid performance, by determining the weight of a plurality of characteristic parameters related to pavement anti-skid performance extracted, multiple characteristic parameters are weighted and handled, and environmental penalty factor is introduced to correct the weighted processing result, obtain anti-skid performance index, so as to utilize anti-skid performance index to evaluate the anti-skid performance of target pavement, to improve the accuracy and evaluation efficiency of pavement anti-skid performance evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pavement engineering, and particularly relates to a super-high-speed three-dimensional laser scanning pavement anti-skid performance evaluation method and system. BACKGROUND

[0002] Pavement anti-skid performance is one of the core indicators for ensuring road traffic safety, and directly affects the braking distance, steering stability and control safety during vehicle driving. In particular, under adverse weather conditions such as rain, ice and snow, insufficient pavement anti-skid performance can easily cause vehicle skidding, rollover and other traffic accidents. With the continuous growth of highway and urban road mileage in China and the rapid increase in the number of motor vehicles, accurate detection and scientific evaluation of pavement anti-skid performance have become a key requirement for road maintenance management and traffic risk prevention and control, and are also an important technical support for extending pavement service life and reducing maintenance costs.

[0003] At present, the main pavement anti-skid performance evaluation methods at home and abroad are mainly divided into two categories: traditional contact detection method and modern non-contact detection method. The two methods have significant differences in detection principle, efficiency and accuracy, and generally have technical limitations, which are difficult to meet the current demand for high precision and high efficiency in road engineering. SUMMARY

[0004] Therefore, the present application provides a super-high-speed three-dimensional laser scanning pavement anti-skid performance evaluation method and system, which solves the technical problem that the current pavement anti-skid performance evaluation method has technical limitations and is difficult to meet the current demand for high precision and high efficiency in road engineering.

[0005] The first aspect of the present application provides a super-high-speed three-dimensional laser scanning pavement anti-skid performance evaluation method, comprising:

[0006] scanning the target pavement using a three-dimensional laser scanner to obtain pavement three-dimensional point cloud raw data;

[0007] preprocessing the pavement three-dimensional point cloud raw data to obtain preprocessed pavement three-dimensional point cloud data;

[0008] constructing a pavement three-dimensional texture model according to the preprocessed pavement three-dimensional point cloud data; wherein the pavement three-dimensional texture model comprises a micro-texture sub-model and a macro-texture sub-model;

[0009] constructing a reference surface of the target pavement and projecting the reference surface into the pavement three-dimensional texture model to obtain a projected pavement three-dimensional texture model;

[0010] extracting a plurality of feature parameters related to pavement anti-skid performance based on the projected pavement three-dimensional texture model;

[0011] Determine the weight of the extracted multiple feature parameters related to the road skid resistance performance, weight the multiple feature parameters according to the weight, introduce an environmental penalty factor to correct the weighted processing result, and obtain a skid resistance performance index;

[0012] According to the skid resistance performance index, the skid resistance performance of the target road is evaluated, and the evaluation result of the skid resistance performance of the target road is obtained.

[0013] Preferably, the preprocessing includes noise removal and outlier rejection.

[0014] Preferably, the road three-dimensional texture model is constructed according to the preprocessed road three-dimensional point cloud data, comprising:

[0015] The preprocessed road three-dimensional point cloud data is gridded according to a predetermined grid size, and a plurality of grid units are obtained;

[0016] For each grid unit, the mean value of the coordinates of all point clouds in the grid unit is calculated, and the mean value of the coordinates is taken as the height value of the grid unit, and a road height matrix is obtained;

[0017] The road height matrix is decomposed into a plurality of sub-matrices of different scales by wavelet transform, and the texture change gradient and texture wavelength of each sub-matrix are determined;

[0018] The texture change gradient of each sub-matrix is compared with a preset gradient threshold to obtain a first comparison result, and the texture wavelength of each sub-matrix is compared with a preset texture wavelength threshold to obtain a second comparison result;

[0019] According to the first comparison result and the second comparison result, all sub-matrices are divided into high-frequency sub-matrices and low-frequency sub-matrices;

[0020] The high-frequency sub-matrices and the low-frequency sub-matrices are respectively inverse wavelet transformed and reconstructed to obtain the micro-texture sub-model and the macro-texture sub-model; wherein the micro-texture sub-model reflects the small concave-convex features of the road surface, and the macro-texture sub-model reflects the large-scale undulation features of the road surface as a whole.

[0021] The micro-texture sub-model is superimposed into the corresponding coordinate position in the macro-texture sub-model to obtain the road three-dimensional texture model.

[0022] Preferably, according to the first comparison result and the second comparison result, all sub-matrices are divided into high-frequency sub-matrices and low-frequency sub-matrices, comprising:

[0023] For each of the sub-matrices, the sub-matrix whose texture change gradient is greater than a preset gradient threshold is divided into a first high-frequency sub-matrix, and the sub-matrix whose texture change gradient is not greater than the preset gradient threshold is divided into a first low-frequency sub-matrix;

[0024] It is judged whether the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is less than a preset texture wavelength threshold, and if the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is not less than the preset texture wavelength threshold, the sub-matrix in the first high-frequency sub-matrix is divided into the first low-frequency sub-matrix, to obtain a second high-frequency sub-matrix and a second low-frequency sub-matrix;

[0025] It is judged whether the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is less than a preset texture wavelength threshold, and if the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is not less than the preset texture wavelength threshold, the preset gradient threshold is reduced, and based on the reduced preset gradient threshold, the comparison of the texture change gradient of each of the sub-matrices with the preset gradient threshold is performed to obtain a first comparison result, and the comparison of the texture wavelength of each of the sub-matrices with the preset texture wavelength threshold is performed to obtain a second comparison result, until all sub-matrices are divided into high-frequency sub-matrices or low-frequency sub-matrices as required.

[0026] Preferably, the reference surface of the target road surface is constructed, and the reference surface is projected into the road surface three-dimensional texture model to obtain a projected road surface three-dimensional texture model, comprising:

[0027] The least square method is used to perform plane fitting on a plurality of preprocessed road surface three-dimensional point cloud data to construct the reference surface of the target road surface;

[0028] The reference surface of the target road surface is projected into the road surface three-dimensional texture model in a perpendicular projection manner to obtain an initial projected road surface three-dimensional texture model;

[0029] The initial projected road surface three-dimensional texture model is subjected to coordinate correction to obtain the projected road surface three-dimensional texture model.

[0030] Preferably, the characteristic parameters include a friction area characteristic parameter;

[0031] The plurality of characteristic parameters related to the road surface anti-skid performance are extracted based on the projected road surface three-dimensional texture model, comprising:

[0032] The projected road surface three-dimensional texture model is subjected to grid division to obtain a plurality of grids;

[0033] For each of the grids, the point cloud corresponding to the average value of the Z-axis coordinates of all point clouds in the grid is extracted as the feature point cloud of the grid;

[0034] For each of the feature point clouds, the texture height of the feature point cloud relative to the reference surface of the target road surface is determined.

[0035] The texture height of each of the feature point clouds is compared with a preset height threshold, and the feature point cloud whose texture height is greater than the preset height threshold is screened out, and the grid corresponding to the screened feature point cloud is determined as the effective friction contact area grid.

[0036] The effective friction contact area is determined according to the number of the effective friction contact area grids and the unit grid area.

[0037] According to the proportion of the effective friction contact area relative to the total area corresponding to all grids, the effective friction contact area proportion is determined as the friction area characteristic parameter.

[0038] Preferably, the characteristic parameters further include a texture morphology parameter; the texture morphology parameter includes an average texture depth, a texture sharpness and a friction area distribution entropy.

[0039] Based on the projected road surface three-dimensional texture model, a plurality of feature parameters related to road surface anti-skid performance are extracted, including:

[0040] According to the texture height of each of the feature point clouds, the average texture depth of all the feature point clouds is determined; wherein the average texture depth is the average value of the texture heights of all the feature point clouds.

[0041] According to the texture heights of all the feature point clouds, the texture height standard deviation is determined, and the ratio of the texture height standard deviation to the average texture depth is taken as the texture sharpness.

[0042] The projected road surface three-dimensional texture model is divided into a plurality of sub-regions according to a preset size, and for each of the sub-regions, the effective friction contact area proportion of each of the sub-regions is calculated.

[0043] According to the effective friction contact area proportions of each of the sub-regions, the friction area distribution entropy is determined; wherein the friction area distribution entropy is used to represent the uniformity of the distribution of the road surface texture in space.

[0044] Preferably, the weight of the extracted plurality of feature parameters related to road surface anti-skid performance is determined, the plurality of feature parameters are weighted according to the weight, and an environmental penalty factor is introduced to modify the weighted processing result to obtain an anti-skid performance index, including:

[0045] standardize each feature parameter to obtain a standardized feature parameter;

[0046] determine a grey correlation degree and a Pearson correlation coefficient between each standardized feature parameter and a reference feature parameter;

[0047] normalize the grey correlation degree and the Pearson correlation coefficient respectively, and linearly fuse the normalized grey correlation degree and the normalized Pearson correlation coefficient to obtain a weight of each feature parameter;

[0048] weight each standardized feature parameter according to the weight of each feature parameter to obtain an initial anti-skid performance index;

[0049] obtain a road surface water content and a road service life of the target road surface, and determine a road surface humidity correction coefficient and a service life correction coefficient according to the road surface water content and the road service life respectively;

[0050] determine the environmental penalty factor according to the road surface humidity correction coefficient and the service life correction coefficient;

[0051] correct the initial anti-skid performance index by the environmental penalty factor to obtain the anti-skid performance index.

[0052] Preferably, the anti-skid performance of the target road surface is evaluated according to the anti-skid performance index to obtain an evaluation result of the anti-skid performance of the target road surface, which comprises:

[0053] obtain a plurality of anti-skid performance thresholds of different anti-skid performance levels, compare the anti-skid performance index with the plurality of anti-skid performance thresholds, and determine an anti-skid performance level to which the target road surface belongs according to the comparison result; wherein the higher the anti-skid performance level is, the higher the anti-skid performance of the target road surface is.

[0054] In a second aspect, the present application further provides a super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation system, comprising:

[0055] a point cloud acquisition module for scanning a target road surface by a three-dimensional laser scanner to obtain road surface three-dimensional point cloud original data;

[0056] a preprocessing module for preprocessing the road surface three-dimensional point cloud original data to obtain preprocessed road surface three-dimensional point cloud data;

[0057] a texture model construction module for constructing a road surface three-dimensional texture model according to the preprocessed road surface three-dimensional point cloud data; wherein the road surface three-dimensional texture model comprises a micro-texture sub-model and a macro-texture sub-model;

[0058] a texture model projection module configured to construct a reference surface of the target pavement and project the reference surface into the three-dimensional pavement texture model to obtain a projected three-dimensional pavement texture model;

[0059] a characteristic parameter extraction module configured to extract a plurality of characteristic parameters related to pavement skid resistance based on the projected three-dimensional pavement texture model;

[0060] a skid resistance index determination module configured to determine weights of the plurality of characteristic parameters related to pavement skid resistance, perform weighted processing on the plurality of characteristic parameters according to the weights, and introduce an environmental penalty factor to correct the weighted processing result to obtain a skid resistance index;

[0061] a skid resistance performance evaluation module configured to evaluate the skid resistance performance of the target pavement according to the skid resistance index to obtain an evaluation result of the skid resistance performance of the target pavement.

[0062] As can be seen from the above technical solutions, the three-dimensional laser scanner is used to scan the target pavement to obtain three-dimensional point cloud raw data of the pavement, the three-dimensional point cloud raw data of the pavement is preprocessed, the three-dimensional pavement texture model is constructed according to the preprocessed three-dimensional point cloud data of the pavement, the model comprehensively includes the micro-texture sub-model and the macro-texture sub-model, the reference surface of the target pavement is constructed, the reference surface is projected into the three-dimensional pavement texture model, the plurality of characteristic parameters related to pavement skid resistance are extracted based on the projected three-dimensional pavement texture model, the weights of the plurality of characteristic parameters related to pavement skid resistance are determined, the plurality of characteristic parameters are weighted processed, the environmental penalty factor is introduced to correct the weighted processing result to obtain the skid resistance index, and then the skid resistance performance of the target pavement is evaluated by using the skid resistance index, so that the technical adaptability of the pavement skid resistance performance evaluation method is increased, and the accuracy and efficiency of the pavement skid resistance performance evaluation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the following embodiment or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0064] Figure 1 A flow chart of a super-high-speed three-dimensional laser scanning pavement skid resistance performance evaluation method provided by an embodiment of the present application;

[0065] Figure 2 A structural schematic diagram of a super-high-speed three-dimensional laser scanning pavement skid resistance performance evaluation system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order for those skilled in the art to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0067] As shown in the drawings, Figure 1 The embodiment of the present application provides a super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method, which comprises the following steps S1 to S7. Among them:

[0068] Step S1, using a three-dimensional laser scanner to scan the target pavement to obtain three-dimensional point cloud original data of the pavement.

[0069] Among them, the three-dimensional laser scanner can set the scanning resolution to 0.1-1.0mm and the scanning frequency to 50-200kHz based on the accuracy requirement, and adopt the "serpentine reciprocating scanning" mode, and the overlapping rate of adjacent scanning lines is not less than 15% to scan the area of the target pavement for pavement anti-skid performance evaluation.

[0070] In the scanning process, the intersection of the road edge line and the center line at the starting end of the scanning area is taken as the origin to establish a local three-dimensional coordinate system (X axis points to the driving direction along the road center line, Y axis is perpendicular to X axis and points to the outside of the road, and Z axis is perpendicular to the pavement and points to the sky), and the local coordinates are converted into geodetic coordinates through the GPS positioning module to ensure the spatial uniqueness of the data.

[0071] Step S2, pre-processing the three-dimensional point cloud original data of the pavement to obtain pre-processed three-dimensional point cloud data of the pavement.

[0072] Among them, the pre-processing includes noise removal and outlier rejection.

[0073] Noise removal is to use a Gaussian filtering algorithm to smooth the three-dimensional point cloud original data of the pavement to eliminate random noise generated in the scanning process due to device vibration, environmental interference and other factors. Specifically, by setting appropriate Gaussian kernel size and standard deviation, the weighted average of each point and its neighborhood points is obtained, thereby obtaining the smoothed point cloud data.

[0074] Outlier rejection is based on the statistical characteristics of the point cloud data and uses a statistical filtering method. First, the average distance of each point in the point cloud data to its neighborhood points is calculated, and the outlier points with a distance greater than the average distance are removed, and further isolated noise is removed to ensure the accuracy and reliability of the point cloud data.

[0075] Step S3, constructing a road surface three-dimensional texture model according to the pretreated road surface three-dimensional point cloud data; wherein the road surface three-dimensional texture model comprises a micro-texture sub-model and a macro-texture sub-model.

[0076] The road surface three-dimensional texture model is constructed by organically fusing the micro-texture features and the macro-texture features. The micro-texture sub-model reflects the small concave-convex features of the road surface, and the macro-texture sub-model reflects the large-scale undulating features of the road surface as a whole, so that the road surface three-dimensional texture model can accurately reflect the real anti-skid performance of the road surface.

[0077] Step S4, constructing a reference surface of the target road surface, and projecting the reference surface into the road surface three-dimensional texture model to obtain a projected road surface three-dimensional texture model.

[0078] The reference surface of the target road surface is obtained by plane fitting of the pretreated multiple road surface three-dimensional point cloud data by the least square method. The reference surface serves as a reference for subsequent projection and feature extraction, and can ensure the accuracy of the projection and the reliability of the feature parameters. The reference surface can measure the relative height of the road surface texture, i.e. the degree of protrusion / depression of the texture relative to the reference surface.

[0079] By projecting the reference surface into the road surface three-dimensional texture model, the reference surface and the road surface three-dimensional texture model are integrated, which facilitates the subsequent extraction of related feature parameters.

[0080] Step S5, extracting multiple feature parameters related to the anti-skid performance of the road surface based on the projected road surface three-dimensional texture model.

[0081] After obtaining the projected road surface three-dimensional texture model, the reference surface and the point cloud data can be used to accurately extract multiple feature parameters related to the anti-skid performance of the road surface.

[0082] Step S6, determining the weights of the extracted multiple feature parameters related to the anti-skid performance of the road surface, weighting the multiple feature parameters according to the weights, and introducing an environmental penalty factor to correct the weighting result to obtain an anti-skid performance index.

[0083] The environmental penalty factor is used for correction, which comprehensively considers the influence of the water content of the road surface and the service life on the anti-skid performance, and ensures that the evaluation result can truly reflect the anti-skid ability of the road surface under actual use conditions.

[0084] Step S7, evaluating the anti-skid performance of the target road surface according to the anti-skid performance index to obtain an evaluation result of the anti-skid performance of the target road surface.

[0085] The anti-skid performance index is obtained by comparing the anti-skid performance index with preset anti-skid performance level thresholds, and the anti-skid performance level of the target road surface can be clearly divided. For example, four levels of excellent, good, medium and poor can be set, which correspond to different ranges of anti-skid performance indexes, so as to intuitively display the anti-skid performance of the road surface.

[0086] It should be noted that, in the embodiment of the present application, the target road surface is scanned by a three-dimensional laser scanner to obtain road surface three-dimensional point cloud original data, the road surface three-dimensional point cloud original data is preprocessed, a road surface three-dimensional texture model is constructed according to the preprocessed road surface three-dimensional point cloud data, the model integrates the micro-texture sub-model and the macro-texture sub-model, the datum plane of the target road surface is constructed, and the datum plane is projected into the road surface three-dimensional texture model. Based on the projected road surface three-dimensional texture model, a plurality of characteristic parameters related to the anti-skid performance of the road surface are extracted, the weights of the plurality of characteristic parameters related to the anti-skid performance of the road surface are determined, the plurality of characteristic parameters are weighted, and an environmental penalty factor is introduced to correct the weighted processing result to obtain the anti-skid performance index. Thus, the anti-skid performance of the target road surface is evaluated by using the anti-skid performance index, thereby increasing the technical adaptability of the road surface anti-skid performance evaluation method and improving the accuracy and efficiency of the road surface anti-skid performance evaluation.

[0087] In some embodiments, constructing a road surface three-dimensional texture model according to the preprocessed road surface three-dimensional point cloud data comprises:

[0088] Step S301, the preprocessed road surface three-dimensional point cloud data is grid processed according to a predetermined grid size to obtain a plurality of grid units.

[0089] The grid size can be in the range of 1mmx1mm to 5mmx5mm. By reasonably setting the grid size, the model accuracy and the calculation efficiency can be balanced, that is, the micro-texture features can be fully captured, and the processing burden caused by excessive data amount can be avoided. Specifically, the grid size can be dynamically adjusted according to the actual scanning resolution and the roughness of the road surface. For example, a larger grid (such as 5mmx5mm) is used for rough road surface, and a smaller grid (such as 1mmx1mm) is used for smooth road surface.

[0090] Step S302, for each grid unit, the mean value of the coordinates of all point clouds in the grid unit is calculated, and the mean value of the coordinates is taken as the height value of the grid unit to obtain a road surface height matrix.

[0091] The road surface height matrix is denoted as H, the dimension of H is MxN, M is the number of rows of grid units, N is the number of columns of grid units, and the elements in the road surface height matrix are the height values of the grid units.

[0092] Step S303: Use wavelet transform to decompose the road surface height matrix into multiple sub-matrices of different scales, and determine the texture change gradient and texture wavelength of each sub-matrix.

[0093] The number of decomposition scales for wavelet transform is determined based on the road surface type. Specifically, the number of decomposition scales for wavelet transform is 5 for asphalt pavement, 3 for cement pavement, and 4 for composite asphalt and cement pavement. Biorthogonal wavelets are selected as the wavelet basis functions, and multi-level biorthogonal wavelet transforms are performed on the road height matrix to obtain multiple sets of sub-matrices at different scales, denoted as... ,in, These are the horizontal, vertical, and diagonal detail submatrices of the k-th layer, respectively, where k is the decomposition layer to which this submatrix belongs.

[0094] The texture change gradient of a submatrix quantifies the intensity of texture fluctuations within the submatrix. The more intense the fluctuations, the larger the texture change gradient, corresponding to high-frequency textures (microscopic); the gentler the fluctuations, the smaller the texture change gradient, corresponding to low-frequency textures (macroscopic).

[0095] The calculation process for the texture change gradient of the submatrix is ​​as follows:

[0096] For each submatrix, compute each element w. i,j (i=1,2,...,p;j=1,2,...,q, The two-dimensional gradient magnitude G i,j This reflects the rate of elevation change of this element relative to its neighboring elements:

[0097]

[0098] In the formula, The actual spatial sampling interval corresponding to the k-th layer submatrix. s is the side length of the grid cell; w i,j Let w be the element in the i-th row and j-th column of the submatrix. i+1,j Let w be the element in the (i+1)th row and jth column of the submatrix. i,j+1 Let be the element in the i-th row and j+1-th column of the submatrix.

[0099] For each submatrix, calculate the weighted average of the gradient magnitudes of all elements as the texture change gradient of that submatrix, with the weights being the squares of the gradient magnitudes. Therefore, the texture change gradient of the submatrix is:

[0100]

[0101] In the formula, Texture variation gradient of the submatrix.

[0102] Texture wavelength is:

[0103] wherein, is the texture wavelength, is the spatial frequency corresponding to the main peak of the power spectral density function obtained after two-dimensional Fourier transform of the sub-matrix, reflecting the frequency of the most prominent texture in the sub-matrix.

[0104] wherein, the texture wavelength can reflect the spatial scale of the most prominent texture in the sub-matrix, the longer the texture wavelength, the larger the scale of the texture, corresponding to the macro-texture feature; the shorter the texture wavelength, the smaller the scale of the texture, corresponding to the micro-texture feature. By calculating the texture variation gradient and the texture wavelength of each sub-matrix, the feature information of the micro-texture and the macro-texture of the road surface can be extracted respectively.

[0105] Step S304, comparing the texture variation gradient of each sub-matrix with the preset gradient threshold to obtain a first comparison result, and comparing the texture wavelength of each sub-matrix with the preset texture wavelength threshold to obtain a second comparison result.

[0106] Step S305, dividing all the sub-matrices into high-frequency sub-matrices and low-frequency sub-matrices according to the first comparison result and the second comparison result.

[0107] wherein, the gradient threshold can be the mean value of the texture variation gradients of multiple sub-matrices at the initial time, and the texture wavelength threshold is the micro-macro texture demarcation wavelength, which is 5mm by default, or based on the actual situation for empirical trend.

[0108] Specifically, dividing all the sub-matrices into high-frequency sub-matrices and low-frequency sub-matrices according to the first comparison result and the second comparison result includes:

[0109] Step S3041, for each sub-matrix, dividing the sub-matrix whose texture variation gradient is greater than the preset gradient threshold into a first high-frequency sub-matrix, and dividing the sub-matrix whose texture variation gradient is not greater than the preset gradient threshold into a first low-frequency sub-matrix.

[0110] wherein, at the level of the texture variation gradient, the sub-matrices are initially divided, the sub-matrices with sharp texture variation are classified into the high-frequency category, reflecting the micro-unevenness characteristics of the road surface; and the sub-matrices with gentle texture variation are classified into the low-frequency category, reflecting the macro-undulation characteristics of the road surface.

[0111] Step S3042, judging whether the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is less than the preset texture wavelength threshold, if the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is not less than the preset texture wavelength threshold, the sub-matrix in the first high-frequency sub-matrix is divided into the first low-frequency sub-matrix, to obtain a second high-frequency sub-matrix and a second low-frequency sub-matrix.

[0112] The first high-frequency sub-matrix and the first low-frequency sub-matrix are subjected to secondary screening and adjustment at the level of the texture wavelength. By setting a texture wavelength threshold, the sub-matrix originally classified as high frequency but having a longer texture wavelength is reclassified to the low frequency category, so that the high-frequency sub-matrix more accurately reflects the microscopic texture characteristics, and the low-frequency sub-matrix more accurately reflects the macroscopic texture characteristics. This double screening mechanism effectively improves the accuracy of the pavement texture model construction, providing a more reliable feature parameter basis for subsequent anti-skid performance evaluation.

[0113] In step S3043, it is determined whether the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is less than the preset texture wavelength threshold. If the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is not less than the preset texture wavelength threshold, the preset gradient threshold is reduced, and the texture change gradient of each sub-matrix is compared with the preset gradient threshold to obtain a first comparison result, and the texture wavelength of each sub-matrix is compared with the preset texture wavelength threshold to obtain a second comparison result, until all sub-matrices are classified into high-frequency sub-matrices or low-frequency sub-matrices according to the requirements.

[0114] In the second low-frequency sub-matrix, for the sub-matrix whose texture wavelength does not satisfy the preset threshold, the gradient threshold is dynamically adjusted and the comparison process is re-executed to realize the iterative optimization of the sub-matrix category.

[0115] Specifically, the gradient threshold is reduced by 10%-15% (for example, from the initial mean value of 0.8 mm to 0.7 mm) each time, and the comparison relationship between the texture change gradient and the wavelength of all sub-matrices is recalculated until the texture wavelength of all sub-matrices in the second low-frequency sub-matrix is less than the preset threshold, or the maximum number of iterations (such as 20 times) is reached. This mechanism ensures that the high-frequency sub-matrix is strictly focused on the microscopic texture (wavelength < 5 mm and gradient > threshold), and the low-frequency sub-matrix accurately represents the macroscopic texture (wavelength ≥ 5 mm and gradient ≤ threshold), and finally forms a clear micro-macro texture double-layer classification structure, providing an accurate feature classification basis for subsequent three-dimensional texture model construction.

[0116] In step S306, the high-frequency sub-matrix and the low-frequency sub-matrix are subjected to wavelet inverse transform reconstruction, respectively, to obtain a microscopic texture sub-model and a macroscopic texture sub-model. The microscopic texture sub-model reflects the fine concave-convex features of the pavement surface, and the macroscopic texture sub-model reflects the large-scale undulation features of the pavement as a whole.

[0117] Wherein, when respectively performing inverse wavelet transform on the high-frequency sub-matrix and the low-frequency sub-matrix, a corresponding inverse transform scale needs to be selected according to the decomposition layer to which the sub-matrix belongs. In specific operation, the high-frequency sub-matrix is subjected to inverse biorthogonal wavelet transform of the same layer number as in decomposition, to reconstruct a micro-texture sub-model, which restores millimeter-level concave-convex features of the pavement through high-frequency detail coefficients, such as texture details with a wavelength <5mm, such as stone edges and small depressions; and the low-frequency sub-matrix is subjected to low-frequency approximation coefficient reconstruction to reconstruct a macro-texture sub-model, which reflects meter-level undulation features of the pavement, such as texture trends with a wavelength ≥5mm, such as ruts and pavement undulations.

[0118] Step S307, superimpose the micro-texture sub-model to a corresponding coordinate position in the macro-texture sub-model to obtain a pavement three-dimensional texture model.

[0119] Wherein, in the superimposition process, it is necessary to ensure that the coordinate systems of the micro-texture sub-model and the macro-texture sub-model are aligned to avoid model distortion due to positional offset. Specifically, by recording the spatial coordinates of each grid cell in the original point cloud data, the height values of the grids in the micro-texture sub-model are accurately superimposed on the corresponding coordinates of the macro-texture sub-model, to form a pavement three-dimensional texture model that has both micro details and macro features. Through the fusion of the double-layer structure (micro layer + macro layer), the model not only retains high-frequency features such as stone edges and small cracks, but also reflects low-frequency features such as ruts and waves, providing a complete multi-scale texture description for subsequent anti-skid performance parameter extraction.

[0120] In some embodiments, a reference surface of the target pavement is constructed, and the reference surface is projected into the pavement three-dimensional texture model to obtain a projected pavement three-dimensional texture model, comprising:

[0121] Step S401, perform plane fitting on a plurality of preprocessed pavement three-dimensional point cloud data using the least squares method to construct a reference surface of the target pavement.

[0122] Wherein, the least squares method is used to fit the pavement reference surface, specifically:

[0123]

[0124] In the formula, is the number of point clouds participating in fitting (80% of the total number of preprocessed point clouds), , , is the x, y, z axis coordinates of the mth point cloud, and a, b, c are the coefficients of the reference surface equation.

[0125] By solving the above least squares problem, the reference surface coefficients a, b, and c are obtained, and subsequent texture parameter calculation is all referenced to this reference surface.

[0126] Step S402, project the reference surface of the target road surface into the road surface three-dimensional texture model in a vertical projection manner to obtain an initial projected road surface three-dimensional texture model.

[0127] In the vertical projection process, each point P (x, y, ax+by+c) of the reference surface is replaced by keeping the x and y coordinates unchanged and replacing the z coordinate with the height value of the corresponding position in the road surface three-dimensional texture model to realize the spatial alignment of the reference surface and the texture model. Specifically, for any point P on the reference surface, find the grid cell with the same x and y coordinates in the road surface three-dimensional texture model, and take the height value H of the grid as the z coordinate after projection to generate the projection point P' (x, y, H (x, y)). The overall projection is completed by traversing all points on the reference surface to form the initial projected road surface three-dimensional texture model. The model takes the reference surface as the reference plane, retains the micro-macro double-layer characteristics of the texture model, and eliminates the global tilt error of the original point cloud, thereby providing a standardized geometric reference framework for subsequent anti-skid performance parameter extraction.

[0128] Step S403, coordinate correction is performed on the initial projected road surface three-dimensional texture model to obtain a projected road surface three-dimensional texture model.

[0129] The coordinate correction is a correction for possible local distortion in the projection process. Specifically, by analyzing the height difference distribution between the reference surface and the texture model in the initial projection model, local height abnormalities caused by uneven point cloud density or grid matching error are identified and corrected. The correction method includes: first, calculate the height difference average of each grid cell and its eight neighborhood grids, if the height difference of a certain grid exceeds 200% of the neighborhood average, it is determined as an abnormal point; then, the height of the abnormal grid is recalculated using the height values of the surrounding 8 normal grids by using the bilinear interpolation method to replace the original value; finally, traverse all grids to complete the overall correction to generate the coordinate-corrected road surface three-dimensional texture model. The model eliminates the projection distortion through local smoothing processing, ensures the accurate alignment of the reference surface and the texture model in space, and provides a reliable three-dimensional geometric description for subsequent anti-skid performance parameter extraction.

[0130] In some embodiments, the feature parameters include friction area feature parameters; in this case, based on the projected road surface three-dimensional texture model, a plurality of feature parameters related to the anti-skid performance of the road surface are extracted, including:

[0131] Step S501, divide the projected road surface three-dimensional texture model into a plurality of grids.

[0132] The grid size is determined by a preset scanning resolution. Specifically, the scanning resolution can be set according to actual needs, such as one grid per millimeter or one grid per two millimeters. The entire projected 3D road surface texture model is divided into regular grids, which are evenly distributed in the horizontal and vertical directions, providing the basic units for subsequent calculation of friction area feature parameters. Each grid has a clear boundary and a corresponding spatial coordinate range, which can accurately cover the local area information in the 3D road surface texture model. Therefore, by analyzing each grid, friction area-related features reflecting the road surface's anti-skid performance can be obtained.

[0133] Step S502: For each grid cell, extract the point cloud corresponding to the average Z-axis coordinate of all point clouds within the grid cell as the feature point cloud of the grid cell.

[0134] The average Z-axis coordinate of all point clouds within a grid is obtained by calculating the average height of all point clouds in the vertical direction (Z-axis). The point cloud corresponding to this average value represents the road surface height characteristics of the grid area. The specific calculation process is as follows: First, all point clouds within the grid are traversed, and the Z-axis coordinate value of each point cloud is extracted. Then, all Z-axis coordinate values ​​are summed and divided by the number of point clouds to obtain the average Z-axis coordinate. Finally, a point cloud with this average Z-axis coordinate is selected or generated as a feature point cloud (if no such feature point cloud exists, a virtual point with this coordinate can be generated to represent it). This point cloud is located at the center height of the grid in space and can reflect the average road surface elevation within the grid area. These feature point clouds not only preserve the microscopic undulation information of the road surface 3D texture model but also achieve data regularization and simplification through rasterization processing, facilitating efficient friction area analysis.

[0135] Step S503: For each feature point cloud, determine the texture height of the Z-axis height of the feature point cloud relative to the reference surface of the target road surface.

[0136] The texture height is the difference between the Z-axis height of the feature point cloud and the Z-axis height of the corresponding point cloud on the reference surface of the target road surface. This height reflects the vertical undulation of the grid area containing the feature point cloud relative to the reference surface. The specific calculation is as follows:

[0137]

[0138] In the formula, For the first The Z-axis height of a feature point cloud relative to the texture height of the reference surface of the target road surface. For the first The Z-axis height of the feature point cloud.

[0139] Among them, if >0 means that the point is a convex part of the road surface texture (participates in friction and support), otherwise it is a flat or concave part (does not participate in support, may be a pore, a groove).

[0140] Step S504, compare the texture height of each feature point cloud with the preset height threshold, filter out the feature point cloud whose texture height is greater than the preset height threshold, and determine the grid corresponding to the filtered feature point cloud as the effective friction contact area grid.

[0141] Wherein, the preset height threshold is 0.2-0.5mm, by filtering out the feature point cloud whose texture height is greater than the preset height threshold, the feature point cloud and the grid participating in friction and support can be determined, and the area that can produce effective friction with the tire tread (excluding small protrusions that are too sharp, easy to wear or unable to fit with the tread).

[0142] Step S505, determine the effective friction contact area according to the number of effective friction contact area grids and the unit grid area.

[0143] Wherein, the product of the number of effective friction contact area grids and the unit grid area is the effective friction contact area.

[0144] Step S506, determine the effective friction contact area ratio as the friction area feature parameter according to the proportion of the effective friction contact area relative to the total area corresponding to all grids.

[0145] Wherein, by calculating the ratio of the effective friction contact area to the total area of all grids, the proportion of the effective area actually participating in friction of the road surface can be quantified. This proportion can intuitively reflect the support ability and friction contribution of the road surface texture to the tire, for example, when the effective friction contact area ratio exceeds 60%, it indicates that the road surface has good anti-skid potential; if the ratio is less than 40%, there may be a risk of insufficient anti-skid performance.

[0146] In some embodiments, the feature parameters further include a texture morphology parameter; the texture morphology parameter includes an average texture depth, a texture sharpness, and a friction area distribution entropy.

[0147] Based on the projected road surface three-dimensional texture model, a plurality of feature parameters related to road surface anti-skid performance are extracted, including:

[0148] Step S511, determine the average texture depth of all feature point clouds according to the texture height of each feature point cloud; wherein the average texture depth is the average value of the texture height of all feature point clouds.

[0149] The calculation process of the average texture depth is as follows: first, the texture height data of all feature point clouds are summarized, and the height data reflects the vertical fluctuation degree of each local area of the road surface relative to the reference surface; then, the texture height values of all feature point clouds are added and divided by the total number of feature point clouds, and the result is the average texture depth. The parameter can comprehensively represent the depth of the overall texture of the road surface. The greater the average texture depth, the more significant the concave-convex features of the road surface, which usually means that the tire can produce stronger mechanical interlocking effect when contacting with the road surface, thereby providing more reliable geometric basis for the anti-skid performance.

[0150] In step S512, the texture height standard deviation is determined according to the texture height of all feature point clouds, and the ratio of the texture height standard deviation to the average texture depth is taken as the texture sharpness.

[0151] The calculation process of the texture sharpness is as follows: first, the standard deviation of the texture height data of all feature point clouds is calculated, which reflects the dispersion degree of the texture height of the road surface, that is, the fluctuation range of the texture; then, the calculated texture height standard deviation is divided by the average texture depth, and the obtained ratio is the texture sharpness. The texture sharpness can quantify the steepness of the road surface texture. The greater the ratio, the more sharp protrusions or depressions exist on the road surface, which will cause more intense stress changes during tire rolling, thereby having a significant impact on the anti-skid performance of the road surface.

[0152] In step S513, the projected three-dimensional texture model of the road surface is divided into a plurality of sub-regions according to a preset size, and for each sub-region, the proportion of the effective friction contact area of each sub-region is calculated.

[0153] The projected three-dimensional texture model of the road surface can be divided into UxL sub-regions according to the region size of 0.5m x 0.5m, and the feature point clouds of each sub-region can be screened out and the proportion of the effective friction contact area of each sub-region can be calculated in a similar manner as steps S502 to S506. The specific process is as follows: performing grid division on each sub-region, extracting the feature point cloud, determining the texture height, screening the effective friction contact area grid, and then calculating the effective friction contact area and the proportion of the total area of the sub-region.

[0154] In step S514, the friction area distribution entropy is determined according to the proportion of the effective friction contact area of each sub-region. The friction area distribution entropy is used to represent the uniformity of the spatial distribution of the road surface texture.

[0155] The calculation of the friction area distribution entropy is based on the information entropy theory, which reflects the spatial distribution characteristics of the road surface texture by quantifying the dispersion degree of the proportion of the effective friction contact area of each sub-region. The specific calculation is as follows:

[0156]

[0157] wherein, is the friction area distribution entropy, the effective friction contact area ratio of the u-th row and the v-th column sub-region, U is the number of sub-regions in the row direction, and L is the number of sub-regions in the column direction. l

[0158] In some embodiments, weights of the extracted plurality of feature parameters related to the road anti-skid performance are determined, the plurality of feature parameters are weighted according to the weights, and an environmental penalty factor is introduced to correct the weighted processing result to obtain an anti-skid performance index, including:

[0159] In step S601, each feature parameter is standardized to obtain each standardized feature parameter.

[0160] In the above formula, since the dimensions of the extracted feature parameters are different, the feature parameters need to be standardized to obtain dimensionless feature parameters, and the dimensionless feature parameters are between 0 and 1.

[0161] In step S602, the grey correlation degree and the Pearson correlation coefficient of each standardized feature parameter and a reference feature parameter are determined.

[0162] In the above formula, the reference feature parameter is selected by the feature parameter corresponding to the historical road anti-skid performance index. The similarity between each standardized feature parameter and the reference feature parameter can be determined by grey correlation analysis, and the higher the correlation degree, the more significant the influence of the feature on the anti-skid performance. At the same time, the Pearson correlation coefficient is used to measure the linear correlation between each feature parameter and the reference parameter, and the closer the absolute value of the correlation coefficient to 1, the stronger the positive / negative correlation between the two. In the specific calculation, a standardized feature parameter matrix and a reference parameter vector are first constructed, and the grey correlation degree value and the Pearson correlation coefficient value of each feature parameter and the reference parameter are calculated. These calculation results will be important basis for subsequent weight allocation. For example, when the grey correlation degree of a certain feature parameter is above 0.8 and the absolute value of the Pearson correlation coefficient exceeds 0.6, it can be determined that there is a strong correlation between the feature and the anti-skid performance, and higher weight should be given in the weight allocation. The specific grey correlation degree and the Pearson correlation coefficient are conventional techniques, and will not be described here.

[0163] In step S603, the grey correlation degree and the Pearson correlation coefficient are normalized respectively, and the normalized grey correlation degree and the normalized Pearson correlation coefficient are linearly fused to obtain the weight of each feature parameter.

[0164] In the above formula, the linear fusion process is:

[0165] ​

[0166] wherein, is a weight, is a weight coefficient in linear fusion, , are the normalized grey correlation and the normalized Pearson correlation coefficient, respectively.

[0167] Step S604, the feature parameters after the standard are weighted and processed according to the weight of each feature parameter, and an initial anti-skid performance index is obtained.

[0168] Step S605, the road surface water content and the road service life of the target road are obtained, and the road surface humidity correction coefficient and the service life correction coefficient are determined according to the road surface water content and the road service life, respectively.

[0169] The road surface water content is obtained by real-time measurement of the humidity sensor at the preset monitoring point position of the target road, and the average road surface water content is obtained by averaging the measurement values of each monitoring point. According to experimental data and engineering experience, a mapping relationship table of road surface water content and humidity correction coefficient is established, and different humidity correction coefficient values are obtained when the road surface average water content is in different intervals. For example, when the water content is less than 5%, the humidity correction coefficient is 1.0; when the water content is between 5%-10%, the humidity correction coefficient is 0.9; when the water content is higher than 10%, the humidity correction coefficient is 0.8. The road service life is obtained by referring to the road construction archives or the records of the relevant management department, and the service life correction coefficient is set by using the exponential decay model according to the service life and the road performance decay law. Generally, the service life correction coefficient is:

[0170]

[0171] wherein, is a service life correction coefficient, is a decay coefficient, and the value is 0.3-0.5, is a service life.

[0172] Step S606, the environmental penalty factor is determined according to the road surface humidity correction coefficient and the service life correction coefficient.

[0173] The calculation method of the environmental penalty factor is: multiplying the road surface humidity correction coefficient and the service life correction coefficient, and the product is the environmental penalty factor.

[0174] Step S607, the initial anti-skid performance index is corrected by the environmental penalty factor, and the anti-skid performance index is obtained.

[0175] The environmental penalty factor and the initial anti-skid performance index are multiplied, and the initial anti-skid performance index is corrected, and the anti-skid performance index is obtained.

[0176] In some embodiments, the anti-skid performance of the target road surface is evaluated according to the anti-skid performance index, and an evaluation result of the anti-skid performance of the target road surface is obtained, including:

[0177] The anti-skid performance threshold values of a plurality of anti-skid performance levels are obtained, the anti-skid performance index is compared with the plurality of anti-skid performance threshold values, and the anti-skid performance level to which the target road surface belongs is determined according to the comparison result; wherein the higher the anti-skid performance level, the higher the anti-skid performance of the target road surface.

[0178] The anti-skid performance level can include excellent, good, medium, and poor, and different anti-skid performance threshold values are set for each anti-skid performance level. For example, when the anti-skid performance level is excellent, the lower limit of the corresponding anti-skid performance threshold value can be set to 0.85; when the level is good, the threshold value range is set to 0.7-0.85; when the level is medium, the threshold value range is set to 0.55-0.7; and when the level is poor, the upper limit of the threshold value is set to 0.55. After the anti-skid performance index is normalized, the normalized anti-skid performance index calculated is compared with the threshold values of each level, and the anti-skid performance level to which the target road surface belongs can be determined. For example, when the anti-skid performance index is 0.88, it is determined to be excellent; and when the index is 0.65, it is determined to be medium. This grading evaluation method can intuitively reflect the advantages and disadvantages of the anti-skid performance of the road surface, and provide a quantitative basis for road maintenance decision-making.

[0179] Based on the same inventive concept, the embodiments of the present application also provide a super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation system for implementing the super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation method.

[0180] The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method, so the specific limitations in one or more super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation system embodiments provided below can refer to the limitations of the super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation method described above, and will not be repeated here.

[0181] As shown in Figure 2 The embodiments of the present application provide a super-high-speed three-dimensional laser scanning road surface anti-skid performance evaluation system, which includes:

[0182] The point cloud acquisition module 100 is used to scan the target road surface by using a three-dimensional laser scanner, and obtain road surface three-dimensional point cloud original data;

[0183] The preprocessing module 200 is used to preprocess the road surface three-dimensional point cloud original data, and obtain preprocessed road surface three-dimensional point cloud data;

[0184] The texture model construction module 300 is configured to construct a road surface three-dimensional texture model according to the preprocessed road surface three-dimensional point cloud data; wherein the road surface three-dimensional texture model comprises a micro-texture sub-model and a macro-texture sub-model.

[0185] The texture model projection module 400 is configured to construct a reference surface of the target road surface, and project the reference surface into the road surface three-dimensional texture model to obtain a projected road surface three-dimensional texture model.

[0186] The characteristic parameter extraction module 500 is configured to extract a plurality of characteristic parameters related to the road surface anti-skid performance based on the projected road surface three-dimensional texture model.

[0187] The anti-skid index determination module 600 is configured to determine the weight of the plurality of characteristic parameters related to the road surface anti-skid performance, perform weighted processing on the plurality of characteristic parameters according to the weight, and introduce an environmental penalty factor to correct the weighted processing result to obtain an anti-skid performance index.

[0188] The anti-skid performance evaluation module 700 is configured to evaluate the anti-skid performance of the target road surface according to the anti-skid performance index to obtain an evaluation result of the anti-skid performance of the target road surface.

[0189] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0190] It should be understood that although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0191] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the division of the units is only a logical function division, and there can be another division manner in actual implementation; for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0192] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent ones; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the anti-skid performance of a superhigh-speed three-dimensional laser scanned pavement, characterized in that, The method comprises the following steps: scanning a target road surface by using a three-dimensional laser scanner to obtain original three-dimensional point cloud data of the road surface; preprocessing the original three-dimensional point cloud data of the road surface to obtain preprocessed three-dimensional point cloud data of the road surface; constructing a three-dimensional texture model of the road surface according to the preprocessed three-dimensional point cloud data of the road surface; wherein the three-dimensional texture model of the road surface comprises a micro-texture sub-model and a macro-texture sub-model, which comprises: performing grid processing on the preprocessed three-dimensional point cloud data of the road surface according to a predetermined grid size to obtain a plurality of grid units; for each grid unit, calculating the mean value of the coordinates of all point clouds in the grid unit, and taking the mean value of the coordinates as the height value of the grid unit to obtain a road surface height matrix; performing multi-scale decomposition on the road surface height matrix by using wavelet transform to obtain a plurality of sub-matrices of different scales, and determining the texture change gradient and texture wavelength of each sub-matrix; comparing the texture change gradient of each sub-matrix with a preset gradient threshold to obtain a first comparison result, and comparing the texture wavelength of each sub-matrix with a preset texture wavelength threshold to obtain a second comparison result; dividing all sub-matrices into high-frequency sub-matrices and low-frequency sub-matrices according to the first comparison result and the second comparison result; respectively performing inverse wavelet transform reconstruction on the high-frequency sub-matrices and the low-frequency sub-matrices to obtain the micro-texture sub-model and the macro-texture sub-model; wherein the micro-texture sub-model reflects the fine concave-convex features of the road surface, and the macro-texture sub-model reflects the large-scale undulation features of the road surface as a whole; superimposing the micro-texture sub-model to the corresponding coordinate position in the macro-texture sub-model to obtain the three-dimensional texture model of the road surface; constructing a reference surface of the target road surface, and projecting the reference surface into the three-dimensional texture model of the road surface to obtain a projected three-dimensional texture model of the road surface; extracting a plurality of feature parameters related to the anti-skid performance of the road surface based on the projected three-dimensional texture model of the road surface; the feature parameters comprise a friction area feature parameter; the method of extracting a plurality of feature parameters related to the anti-skid performance of the road surface based on the projected three-dimensional texture model of the road surface comprises: dividing the projected three-dimensional texture model of the road surface into a plurality of grids; for each grid, extracting the point cloud corresponding to the average value of the Z-axis coordinates of all point clouds in the grid as the feature point cloud of the grid; for each feature point cloud, determining the texture height of the feature point cloud relative to the texture height of the reference surface of the target road surface; comparing the texture height of each feature point cloud with a preset height threshold, screening out the feature point cloud whose texture height is greater than the preset height threshold, and determining the grid corresponding to the screened feature point cloud as an effective friction contact area grid; determining the effective friction contact area based on the number of effective friction contact area grids and the area of a unit grid; determining the effective friction contact area ratio as the friction area feature parameter based on the proportion of the effective friction contact area relative to the total area corresponding to all grids. The characteristic parameters further include a texture form parameter; the texture form parameter includes an average texture depth, a texture sharpness, and a friction area distribution entropy; The method further includes: According to the texture heights of each of the feature point clouds, determining an average texture depth of all the feature point clouds; wherein the average texture depth is an average value of the texture heights of all the feature point clouds; According to the texture heights of all the feature point clouds, determining a texture height standard deviation, and taking a ratio of the texture height standard deviation to the average texture depth as the texture sharpness; Dividing the projected road three-dimensional texture model into a plurality of sub-regions according to a preset size, and calculating, for each of the sub-regions, an effective friction contact area proportion of each of the sub-regions; According to the effective friction contact area proportions of the sub-regions, determining a friction area distribution entropy; wherein the friction area distribution entropy is used to represent a uniformity of distribution of the road texture in space; Determining weights of the extracted plurality of characteristic parameters related to the road anti-skid performance, performing weighted processing on the plurality of characteristic parameters according to the weights, and introducing an environmental penalty factor to correct the weighted processing result to obtain an anti-skid performance index; According to the anti-skid performance index, evaluating the anti-skid performance of the target road to obtain an evaluation result of the anti-skid performance of the target road.

2. The super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method according to claim 1, characterized in that, The preprocessing includes noise removal and outlier elimination.

3. The super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method according to claim 1, characterized in that, The dividing all the sub-matrices into high-frequency sub-matrices and low-frequency sub-matrices according to the first comparison result and the second comparison result includes: For each of the sub-matrices, dividing a sub-matrix with a texture change gradient greater than a preset gradient threshold into a first high-frequency sub-matrix, and dividing a sub-matrix with a texture change gradient not greater than the preset gradient threshold into a first low-frequency sub-matrix; Determining whether the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is less than a preset texture wavelength threshold, and if the texture wavelength of the sub-matrix in the first high-frequency sub-matrix is not less than the preset texture wavelength threshold, dividing the sub-matrix in the first high-frequency sub-matrix into the first low-frequency sub-matrix to obtain a second high-frequency sub-matrix and a second low-frequency sub-matrix; Determining whether the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is less than the preset texture wavelength threshold, and if the texture wavelength of the sub-matrix in the second low-frequency sub-matrix is not less than the preset texture wavelength threshold, reducing the preset gradient threshold, and based on the reduced preset gradient threshold, returning to the comparing the texture change gradient of each of the sub-matrices with the preset gradient threshold to obtain a first comparison result, and comparing the texture wavelength of each of the sub-matrices with the preset texture wavelength threshold to obtain a second comparison result, until all the sub-matrices are divided into high-frequency sub-matrices or low-frequency sub-matrices as required.

4. The super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method according to claim 1, characterized in that, The constructing a reference surface of the target road and projecting the reference surface into the road three-dimensional texture model to obtain a projected road three-dimensional texture model includes: The least square method is used to perform plane fitting on multiple preprocessed road three-dimensional point cloud data, and a reference surface of the target road is constructed; The reference surface of the target road is projected into the road three-dimensional texture model in a vertical projection manner to obtain an initial projected road three-dimensional texture model; The initial projected road three-dimensional texture model is subjected to coordinate correction to obtain the projected road three-dimensional texture model.

5. The super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method according to claim 1, characterized in that, The weight of the extracted multiple feature parameters related to the road anti-skid performance is determined, the multiple feature parameters are subjected to weighted processing according to the weight, and an environmental penalty factor is introduced to correct the weighted processing result to obtain an anti-skid performance index, which comprises: The standardization processing is performed on each feature parameter to obtain each standardized feature parameter; The grey correlation degree and the Pearson correlation coefficient of each standardized feature parameter and a reference feature parameter are determined; The grey correlation degree and the Pearson correlation coefficient are subjected to normalization processing respectively, and the normalized grey correlation degree and the normalized Pearson correlation coefficient are subjected to linear fusion to obtain the weight of each feature parameter; The weighted processing is performed on each standardized feature parameter according to the weight of each feature parameter to obtain an initial anti-skid performance index; The road water content and the road service life of the target road are obtained, and a road humidity correction coefficient and a service life correction coefficient are respectively determined according to the road water content and the road service life; The environmental penalty factor is determined according to the road humidity correction coefficient and the service life correction coefficient; The initial anti-skid performance index is corrected by the environmental penalty factor to obtain the anti-skid performance index.

6. The super-speed three-dimensional laser scanning pavement anti-skid performance evaluation method according to claim 1, characterized in that, The anti-skid performance of the target road is evaluated according to the anti-skid performance index to obtain an evaluation result of the anti-skid performance of the target road, which comprises: A plurality of anti-skid performance threshold values of different anti-skid performance levels are obtained, the anti-skid performance index is compared with the plurality of anti-skid performance threshold values, and the anti-skid performance level to which the target road belongs is determined according to the comparison result; wherein the higher the anti-skid performance level is, the higher the anti-skid performance of the target road is.

7. A system for evaluating the anti-skid performance of a road surface by super-high-speed three-dimensional laser scanning, which executes the method for evaluating the anti-skid performance of a road surface by super-high-speed three-dimensional laser scanning according to claim 1, characterized by It comprises: A point cloud acquisition module is used to scan the target road by using a three-dimensional laser scanner to obtain road three-dimensional point cloud original data; A preprocessing module is used to preprocess the road three-dimensional point cloud original data to obtain preprocessed road three-dimensional point cloud data; A texture model construction module is used to construct a road three-dimensional texture model according to the preprocessed road three-dimensional point cloud data; wherein the road three-dimensional texture model comprises a micro-texture sub-model and a macro-texture sub-model; A texture model projection module is used to construct a reference surface of the target road and project the reference surface into the road three-dimensional texture model to obtain a projected road three-dimensional texture model; A feature parameter extraction module is used to extract multiple feature parameters related to the road anti-skid performance based on the projected road three-dimensional texture model; The anti-skid index determination module is configured to determine weights of the extracted characteristic parameters related to the anti-skid performance of the road surface, perform weighted processing on the characteristic parameters according to the weights, introduce an environmental penalty factor to correct the weighted processing result, and obtain an anti-skid performance index; The anti-skid performance evaluation module is configured to evaluate the anti-skid performance of the target road surface according to the anti-skid performance index, and obtain an evaluation result of the anti-skid performance of the target road surface.

Images