Roughness classification method based on precise three dimensions

Abstract

The present application provides a kind of based on precision three-dimensional rut classification method, comprising: obtaining road elevation data and road grayscale data;Based on the road elevation data and road grayscale data, obtain the cross section control profile in lane range;The upper envelope line of the cross section control profile is obtained, and the cross section ideal control profile is obtained in combination with the cross section control profile;Based on the cross section control profile and cross section ideal control profile, the deformation area of cross section is divided, to obtain multiple cross section deformation areas;Based on the cross section control profile, cross section ideal control profile and multiple cross section deformation areas, obtain cross section deformation area parameter;Based on the multiple cross section deformation area parameters, the rut is classified.Use to solve the defect that rut category is difficult to distinguish in the prior art, realize the accurate classification of rut.

Description

Technical Field

[0001] This invention relates to the field of road inspection technology, and in particular to a method for rut classification based on precise three-dimensional analysis. Background Technology

[0002] Rutting, as one of the major defects of asphalt pavement, directly affects driving safety and comfort. Severe rutting can also damage road structure and shorten highway lifespan. Studies show that rutting type is closely related to the layer of road damage, and different treatment methods should be adopted for different types of rutting. Therefore, accurate rutting detection and classification are of great guiding significance for road performance evaluation and maintenance plan development.

[0003] Current rutting detection methods calculate rutting depth based on discrete cross-sectional elevation measurements. Due to the large gaps between the original measurement data, continuous rutting morphological changes cannot be captured, often leading to an underestimation of the maximum rutting depth. Secondly, the same rutting depth value may correspond to several different cross-sectional morphologies, failing to accurately reflect the impact of different types of rutting on the pavement structure. Furthermore, rutting depth cannot reflect the material used in rutting milling, making it difficult to distinguish rutting categories and failing to provide managers with comprehensive, accurate, and reliable rutting information. Summary of the Invention

[0004] This invention provides a precise three-dimensional rut classification method to overcome the shortcomings of existing technologies in distinguishing rut categories and achieve accurate rut classification.

[0005] This invention provides a method for rut classification based on precise three-dimensional analysis, comprising:

[0006] Acquire road surface elevation data and road surface grayscale data;

[0007] Based on the road surface elevation data and road surface grayscale data, the cross-sectional control profile within the lane range is obtained;

[0008] Obtain the upper envelope of the cross-sectional control profile, and combine it with the cross-sectional control profile to obtain the ideal cross-sectional control profile.

[0009] Based on the cross-sectional control profile and the ideal cross-sectional control profile, the cross-section is divided into deformation regions to obtain multiple cross-sectional deformation regions.

[0010] Based on the cross-sectional control profile, the ideal cross-sectional control profile, and multiple cross-sectional deformation regions, the parameters of the cross-sectional deformation regions are obtained.

[0011] The ruts are classified based on the parameters of the deformation regions of the multiple cross sections.

[0012] According to a precise three-dimensional rut classification method provided by the present invention, the step of obtaining the cross-sectional control profile within the lane range based on the road surface elevation data and road surface grayscale data includes:

[0013] Based on the road surface elevation data and road surface grayscale data, obtain the target road surface elevation data and target road surface grayscale data within the lane range;

[0014] Based on any cross section in the target road surface elevation data, obtain the cross section control profile.

[0015] According to a precise three-dimensional rut classification method provided by the present invention, the step of obtaining target road surface elevation data and target road surface grayscale data within the lane range based on the road surface elevation data and road surface grayscale data includes:

[0016] Based on the road surface elevation data, the first region of potential lane lines is marked using the elevation and geometric features of the road lane lines.

[0017] Based on the road surface grayscale data, the reflective properties and geometric dimensions of the road lane lines are used to mark the potential second lane line regions;

[0018] By combining the first potential lane line region and the second potential lane line region, the current lane line position on the road surface is determined;

[0019] Based on the current lane line position, extract the target road surface elevation data and target road surface grayscale data within the lane range.

[0020] According to a precise three-dimensional rut classification method provided by the present invention, the step of obtaining the upper envelope of the cross-sectional control profile and combining it with the cross-sectional control profile to obtain the ideal cross-sectional control profile includes:

[0021] Obtain the upper envelope of the cross-sectional control profile;

[0022] Based on the cross-sectional control profile and its upper envelope, the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross-section are obtained.

[0023] Based on the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross section, the ideal control profile of the cross section is obtained.

[0024] According to a precise three-dimensional rut classification method provided by the present invention, the step of obtaining the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross section based on the cross section control profile and the upper envelope of the cross section control profile includes:

[0025] The cross-sectional control contour is divided into multiple sub-segments according to a preset step size;

[0026] For any field, calculate the average absolute distance from all points of the sub-segment in the cross-sectional control profile to the upper envelope;

[0027] Select n segments whose average absolute distance is less than a preset distance, and calculate the direction of each segment in the n segments through linear fitting.

[0028] Based on the direction of each of the n sub-segments, obtain the representative direction of the sub-segment of the unworn region;

[0029] Among the multiple sub-segments at the left end of the cross section, the sub-segment most similar to the representative direction of the sub-road segment is selected as the unworn area at the left end. The average value of the cross section control profile corresponding to the unworn area at the left end is taken as the representative elevation value of the unworn area at the left end. The average value of the abscissa in the road width direction corresponding to the unworn area at the left end is taken as the representative abscissa position of the unworn area at the left end.

[0030] Among the multiple sub-segments at the right end of the cross section, the sub-segment most similar to the representative direction of the sub-road segment is selected as the right-end unworn area. The average value of the cross section control profile corresponding to the right-end unworn area is taken as the representative elevation value of the right-end unworn area, and the average value of the road width direction corresponding to the right-end unworn area is taken as the representative abscissa position of the right-end unworn area.

[0031] According to a precise three-dimensional rut classification method provided by the present invention, the step of obtaining the ideal control profile of the cross-section based on the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross-section includes:

[0032] The left endpoint is defined by the representative horizontal coordinate position and the representative elevation value corresponding to the unworn area at the left end, and the right endpoint is defined by the representative horizontal coordinate position and the representative elevation value corresponding to the unworn area at the right end.

[0033] Connect the left and right endpoints, and use the line as the ideal control profile of the cross section.

[0034] According to a precise three-dimensional rut classification method provided by the present invention, the cross-section is divided into deformation regions based on the cross-sectional control profile and the ideal cross-sectional control profile to obtain multiple cross-sectional deformation regions, including:

[0035] For any measuring point in the cross-sectional control profile, calculate the difference between the ideal cross-sectional control profile and the cross-sectional control profile to obtain the set of cross-sectional difference values ​​between the cross-sectional control profile and the ideal cross-sectional control profile.

[0036] Based on the set of cross-sectional differences, the cross-section is divided into deformation regions according to the continuity of the differences. Regions with continuously positive or continuously negative differences are divided into one region, resulting in multiple cross-sectional deformation regions.

[0037] According to a precise three-dimensional rut classification method provided by the present invention, the cross-sectional deformation region parameters include: deformation depth, deformation width, and deformation area of ​​the cross-sectional deformation region. The method for obtaining the cross-sectional deformation region parameters based on the cross-sectional control profile, the ideal cross-sectional control profile, and multiple cross-sectional deformation regions includes:

[0038] For any cross-sectional deformation region, calculate the absolute difference between the cross-sectional control profile and the cross-sectional ideal control profile to obtain the set of absolute differences between the cross-sectional control profile and the cross-sectional ideal control profile deformation regions.

[0039] For any cross-sectional deformation region, based on the set of absolute differences in deformation regions, the maximum difference in deformation regions is obtained, and the maximum difference in deformation regions is used as the deformation depth of the current cross-sectional deformation region; based on the deformation depths of the multiple cross-sectional deformation regions, the maximum deformation depth of the cross-section is obtained.

[0040] Calculate the effective deformation depth threshold based on the maximum deformation depth of the cross-section;

[0041] For any cross-sectional deformation region, count the number of regions whose absolute value is greater than the effective deformation depth threshold in the set of absolute differences of deformation regions corresponding to the cross-sectional deformation region, and obtain the deformation width of the cross-sectional deformation region by combining the lateral sampling interval of the precise three-dimensional point cloud data.

[0042] For any cross-sectional deformation region, the deformation area of ​​the cross-sectional deformation region is calculated based on the deformation depth and deformation width of the cross-sectional deformation region.

[0043] According to the present invention, a method for classifying ruts based on precise three-dimensional analysis is provided, wherein classifying ruts based on the plurality of cross-sectional deformation region parameters includes:

[0044] For any region among the multiple cross-sectional deformation regions, the average value of the road width direction corresponding to the region is taken as the representative abscissa position of the cross-sectional deformation region.

[0045] Based on the width of the cross section along the road width direction, the cross section is divided into multiple statistical regions;

[0046] For any region among the multiple cross-sectional deformation regions, based on the matching of the representative abscissa position of the cross-sectional deformation region with the position of the statistical region, and the consistency between the deformation direction of the cross-sectional deformation region and the preset deformation direction of the statistical region, the cross-sectional deformation region is classified in the statistical region.

[0047] Based on multiple statistical regions, the cumulative deformation area is statistically analyzed region by region. The deformation depth and deformation width corresponding to the maximum deformation area within the statistical region are obtained, and the deformation ratio is calculated.

[0048] Based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the cross section in the multiple statistical regions, the cross section ruts are classified into structural ruts, unstable ruts, abrasive ruts, or compacted ruts.

[0049] According to a precise three-dimensional rut classification method provided by the present invention, the method classifies cross-sectional ruts into structural ruts, unstable ruts, abrasive ruts, or compacted ruts based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the cross-section of the multiple statistical regions.

[0050] A1: Based on the cumulative deformation area of ​​the multiple statistical regions, the sum of the positive deformation area and the sum of the negative deformation area of ​​the cross section are calculated respectively, and the ratio of the negative deformation area to the positive deformation area is calculated.

[0051] A2: If the ratio of the negative deformation area to the positive deformation area is less than the first preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset first deformation depth threshold, then the cross section rut is classified as a structural rut; otherwise, proceed to step A3.

[0052] A3: If the ratio of the negative deformation area to the positive deformation area is greater than the second preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset second deformation depth threshold, then the cross section rut is classified as an unstable rut; otherwise, proceed to step A4.

[0053] A4: If the deformation ratio of two statistical regions in multiple statistical regions is greater than the preset deformation ratio threshold, and the maximum deformation depth of the cross section is greater than the preset third deformation depth threshold, then the cross section rut is classified as a compacted rut; otherwise, proceed to step A5.

[0054] A5: The cross-sectional ruts are classified as abrasive ruts;

[0055] Wherein, the preset first deformation depth threshold is greater than the preset second deformation depth threshold; the preset second deformation depth threshold is greater than the preset third deformation depth threshold;

[0056] The first preset positive-negative area ratio threshold is less than the second preset positive-negative area ratio threshold.

[0057] This invention provides a precise 3D-based rut classification method. It acquires road surface elevation and grayscale data using a line-scanning 3D measurement sensor to obtain the cross-sectional control profile within the lane range. Based on the upper envelope of the cross-sectional control profile and in conjunction with the profile itself, an ideal cross-sectional control profile is obtained. Then, based on the cross-sectional control profile and the ideal profile, the cross-section is divided into deformation regions, resulting in multiple deformation regions. Parameters for these deformation regions are obtained to reflect the impact of rutting on road surface performance and provide important reference for rut classification. Finally, the ruts are classified based on these parameters. Through these steps, this invention upgrades road surface detection data from two-dimensional to three-dimensional. Based on precise 3D point cloud data of the road surface, it can accurately acquire the profile morphology and effectively distinguish rut categories. Accurate rut classification has significant guiding significance for road performance evaluation and maintenance plan formulation. This invention solves the deficiency of existing technologies in distinguishing rut categories and achieves accurate rut classification. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating the precise three-dimensional rut classification method provided by the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0061] The following is combined Figure 1 This invention describes a rut classification method based on precise three-dimensional analysis.

[0062] like Figure 1 As shown, this embodiment of the invention provides a method for rut classification based on precise three-dimensional modeling, specifically including the following steps (in this embodiment, the numbering of each step is only for distinguishing steps and does not limit the specific execution order of each step):

[0063] Step S1: Obtain road surface elevation data and road surface grayscale data.

[0064] A line-scan 3D measurement sensor simultaneously acquires road surface elevation and grayscale data. The line-scan 3D measurement sensor, mounted on a vehicle-mounted platform, includes a laser and a 3D camera. During measurement, the line-scan 3D measurement sensor continuously collects road surface elevation and grayscale information along the road direction, simultaneously acquiring both data. The acquisition interval for both elevation and grayscale data in the road width direction is ≤5mm, and full lane coverage is required in the road width direction. A single line-scan 3D measurement sensor or multiple sensors can be used.

[0065] Step S2: Based on the road surface elevation data and road surface grayscale data, obtain the cross-sectional control profile within the lane range.

[0066] Based on road surface elevation data and road surface grayscale data, the cross-sectional control profile within the lane area is obtained. The cross-sectional control profile is a low-frequency signal containing macroscopic road surface deformation defects (excluding local texture information), obtained by filtering the cross-section (such as median filtering, low-pass filtering, and mean filtering) or frequency domain transformation (such as Fourier transform and wavelet transform).

[0067] Step S3: Obtain the upper envelope of the cross-sectional control profile, and combine it with the cross-sectional control profile to obtain the ideal cross-sectional control profile.

[0068] The upper envelope above the cross-sectional control contour is obtained by using the envelope extraction method, and combined with the cross-sectional control contour, the ideal cross-sectional control contour is obtained (used to characterize the cross-sectional control contour after removing macroscopic deformation defects from the cross-sectional control contour).

[0069] Step S4: Based on the cross-sectional control profile and the ideal cross-sectional control profile, the cross-section is divided into deformation regions to obtain multiple cross-sectional deformation regions.

[0070] Based on the cross-sectional control profile and the ideal cross-sectional control profile, the cross-section is divided into deformation regions to obtain multiple cross-sectional deformation regions, which are the deformation regions under the condition of road surface deformation, and are used to determine the rutting situation.

[0071] Step S5: Based on the cross-sectional control profile, the ideal cross-sectional control profile, and multiple cross-sectional deformation regions, obtain the cross-sectional deformation region parameters.

[0072] Based on the cross-sectional control profile, the ideal cross-sectional control profile, and multiple cross-sectional deformation regions, cross-sectional deformation region parameters are obtained to provide an accurate reference for rut classification.

[0073] Step S6: Classify the ruts based on the parameters of the multiple cross-sectional deformation regions.

[0074] The parameters obtained from the above steps are analyzed to achieve accurate classification of wheel ruts.

[0075] The present invention provides a precise 3D-based rut classification method. This method acquires road surface elevation and grayscale data using a line-scan 3D measurement sensor to obtain the cross-sectional control profile within the lane range. Based on the upper envelope of the cross-sectional control profile and in conjunction with the cross-sectional control profile, an ideal cross-sectional control profile is obtained. Then, based on the cross-sectional control profile and the ideal cross-sectional control profile, the cross-section is divided into deformation regions, resulting in multiple cross-sectional deformation regions. Parameters of these deformation regions are obtained to reflect the impact of rutting on road surface performance and provide important reference for rut classification. Finally, the ruts are classified based on these multiple cross-sectional deformation region parameters. Through these steps, the present invention upgrades road surface detection data from two-dimensional to three-dimensional. Based on precise 3D point cloud data of the road surface, the contour morphology can be accurately obtained, and rut categories can be effectively distinguished. Accurate rut classification has significant guiding significance for road performance evaluation and maintenance plan formulation. This invention solves the deficiency of existing technologies in distinguishing rut categories and achieves accurate rut classification.

[0076] In this embodiment, obtaining the cross-sectional control profile within the lane range based on the road surface elevation data and road surface grayscale data includes:

[0077] Based on the road surface elevation data and road surface grayscale data, obtain the target road surface elevation data and target road surface grayscale data within the lane range;

[0078] Based on any cross section in the target road surface elevation data, obtain the cross section control profile.

[0079] Based on road surface elevation data and road surface grayscale data, target road surface elevation data and target road surface grayscale data are obtained within the lane range. Then, based on any cross section in the target road surface elevation data, a cross section control profile is obtained, thereby providing an accurate reference for rut classification.

[0080] In this embodiment, obtaining target road surface elevation data and target road surface grayscale data within the lane range based on the road surface elevation data and road surface grayscale data includes:

[0081] Based on the road surface elevation data, the first region of potential lane lines is marked using the elevation and geometric features of the road lane lines.

[0082] Based on the road surface grayscale data, the reflective properties and geometric dimensions of the road lane lines are used to mark the potential second lane line regions;

[0083] By combining the first potential lane line region and the second potential lane line region, the current lane line position on the road surface is determined;

[0084] Based on the current lane line position, extract the target road surface elevation data and target road surface grayscale data within the lane range.

[0085] Based on road surface elevation data, potential lane line areas are marked using the elevation characteristics (lane line elevation is higher than normal road surface areas) and geometric dimensions. Based on road surface grayscale data, potential second lane line areas are marked using the reflective properties (lane line areas have higher grayscale values) and geometric dimensions. Combining the potential first and second lane line areas, the current lane line positions are confirmed. Based on the current lane line positions, target road surface elevation and grayscale data are extracted within the lane area. These steps further refine the target road surface elevation and grayscale data, allowing for further filtering of valid data to ensure its validity and reference value.

[0086] In this embodiment, obtaining the upper envelope of the cross-sectional control profile and combining it with the cross-sectional control profile to obtain the ideal cross-sectional control profile includes:

[0087] Obtain the upper envelope of the cross-sectional control profile;

[0088] Based on the cross-sectional control profile and its upper envelope, the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross-section are obtained.

[0089] Based on the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross section, the ideal control profile of the cross section is obtained.

[0090] Based on the cross-sectional control profile and its upper envelope, representative abscissa positions and elevation values ​​of the unworn areas at both ends of the cross-section are obtained. Based on these representative abscissa positions and elevation values, the ideal control profile of the cross-section is then obtained. Through these steps, typical representative measuring point information of the unworn areas in the cross-sectional profile is obtained to accurately reconstruct the ideal profile of the road surface, thereby correctly reflecting the rut morphology.

[0091] In this embodiment, obtaining the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross section based on the cross section control profile and its upper envelope includes:

[0092] The cross-sectional control contour is divided into multiple sub-segments according to a preset step size;

[0093] For any field, calculate the average absolute distance from all points of the sub-segment in the cross-sectional control profile to the upper envelope;

[0094] Select n segments whose average absolute distance is less than a preset distance, and calculate the direction of each segment in the n segments through linear fitting.

[0095] Based on the direction of each of the n sub-segments, obtain the representative direction of the sub-segment of the unworn region;

[0096] Among the multiple sub-segments at the left end of the cross section, the sub-segment most similar to the representative direction of the sub-road segment is selected as the unworn area at the left end. The average value of the cross section control profile corresponding to the unworn area at the left end is taken as the representative elevation value of the unworn area at the left end. The average value of the abscissa in the road width direction corresponding to the unworn area at the left end is taken as the representative abscissa position of the unworn area at the left end.

[0097] Among the multiple sub-segments at the right end of the cross section, the sub-segment most similar to the representative direction of the sub-road segment is selected as the right-end unworn area. The average value of the cross section control profile corresponding to the right-end unworn area is taken as the representative elevation value of the right-end unworn area, and the average value of the road width direction corresponding to the right-end unworn area is taken as the representative abscissa position of the right-end unworn area.

[0098] In the multiple sub-segments at the left end of the cross-section, according to a preset number M L The sub-segment most similar to the representative direction of the sub-road segment is selected as the left-end unworn area. The average value of the cross-sectional control profile corresponding to the left-end unworn area is used as the representative elevation value of the left-end unworn area, and the average abscissa of the road width direction corresponding to the left-end unworn area is used as the representative abscissa position of the left-end unworn area. Among the multiple sub-segments on the right end of the cross-section, a preset number M is selected. R The sub-segment most similar in direction to the sub-segment is selected as the unworn area on the right. The average value of the cross-sectional control profile corresponding to the unworn area on the right is used as the representative elevation value of the unworn area on the right, and the average value of the road width direction corresponding to the unworn area on the right is used as the representative abscissa position of the unworn area on the right. This method is used to obtain the representative abscissa positions and representative elevation values ​​of the unworn areas on both ends of the cross-section, ensuring the accuracy of the ideal control profile acquisition.

[0099] In this embodiment, obtaining the ideal control profile of the cross-section based on the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross-section includes:

[0100] The left endpoint is defined by the representative horizontal coordinate position and the representative elevation value corresponding to the unworn area at the left end, and the right endpoint is defined by the representative horizontal coordinate position and the representative elevation value corresponding to the unworn area at the right end.

[0101] Connect the left and right endpoints, and use the line as the ideal control profile of the cross section.

[0102] By taking the representative horizontal coordinate position and representative elevation value corresponding to the unworn area on the left as the left endpoint and the representative horizontal coordinate position and representative elevation value corresponding to the unworn area on the right as the right endpoint, the corresponding left endpoint and right endpoint are determined, and the connecting line is used as the ideal control profile of the cross section to accurately restore the original shape of the road surface.

[0103] In this embodiment, the cross-section is divided into deformation regions based on the cross-sectional control profile and the ideal cross-sectional control profile, resulting in multiple cross-sectional deformation regions, including:

[0104] For any measuring point in the cross-sectional control profile, calculate the difference between the ideal cross-sectional control profile and the cross-sectional control profile to obtain the set of cross-sectional difference values ​​between the cross-sectional control profile and the ideal cross-sectional control profile.

[0105] Based on the set of cross-sectional differences, the cross-section is divided into deformation regions according to the continuity of the differences. Regions with continuously positive or continuously negative differences are divided into one region, resulting in multiple cross-sectional deformation regions.

[0106] By calculating the set of cross-sectional differences and dividing the cross-section into deformation areas based on the continuity of the differences, the accurate cross-sectional deformation areas can be obtained, so as to accurately restore the rut shape.

[0107] In this embodiment, the cross-sectional deformation region parameters include: deformation depth, deformation width, and deformation area of ​​the cross-sectional deformation region. The step of obtaining the cross-sectional deformation region parameters based on the cross-sectional control contour, the ideal cross-sectional control contour, and multiple cross-sectional deformation regions includes:

[0108] For any cross-sectional deformation region, calculate the absolute difference between the cross-sectional control profile and the cross-sectional ideal control profile to obtain the set of absolute differences between the cross-sectional control profile and the cross-sectional ideal control profile deformation regions.

[0109] For any cross-sectional deformation region, based on the set of absolute differences in deformation regions, the maximum difference in deformation regions is obtained, and the maximum difference in deformation regions is used as the deformation depth of the current cross-sectional deformation region; based on the deformation depths of the multiple cross-sectional deformation regions, the maximum deformation depth of the cross-section is obtained.

[0110] Calculate the effective deformation depth threshold based on the maximum deformation depth of the cross-section;

[0111] For any cross-sectional deformation region, count the number of regions whose absolute value is greater than the effective deformation depth threshold in the set of absolute differences of deformation regions corresponding to the cross-sectional deformation region, and obtain the deformation width of the cross-sectional deformation region by combining the lateral sampling interval of the precise three-dimensional point cloud data.

[0112] For any cross-sectional deformation region, the deformation area of ​​the cross-sectional deformation region is calculated based on the deformation depth and deformation width of the cross-sectional deformation region.

[0113] For any cross-sectional deformation region, calculate the absolute difference between the cross-sectional control profile and the ideal cross-sectional control profile to obtain the set of absolute differences between the cross-sectional control profile and the ideal cross-sectional control profile in the deformation region. Based on the set of absolute differences in the deformation region, obtain the maximum difference D of the deformation region. max The maximum difference D in the deformed region max The deformation depth of the current cross-sectional deformation region is used as the basis for determining the maximum deformation depth D' of the cross-section. max Based on the maximum deformation depth D' of the cross-section max Calculate the effective deformation depth threshold T. Count the number of regions in the set of absolute differences corresponding to the cross-sectional deformation regions whose absolute values ​​are greater than the effective deformation depth threshold T, denoted as u. Combine this with the lateral sampling interval x of the precise 3D point cloud data to obtain the deformation width D of the cross-sectional deformation region. w (D w =u*x). Based on the deformation depth and width of the cross-sectional deformation region, the deformation area of ​​the cross-sectional deformation region is calculated. The deformation depth, deformation width, and deformation area of ​​the cross-sectional deformation region are obtained respectively to accurately obtain the three-dimensional shape data of the contour, thereby effectively distinguishing the rut categories.

[0114] In this embodiment, the classification of ruts based on the multiple cross-sectional deformation region parameters includes:

[0115] For any region among the multiple cross-sectional deformation regions, the average value of the road width direction corresponding to the region is taken as the representative abscissa position of the cross-sectional deformation region.

[0116] Based on the width of the cross section along the road width direction, the cross section is divided into multiple statistical regions;

[0117] For any region among the multiple cross-sectional deformation regions, based on the matching of the representative abscissa position of the cross-sectional deformation region with the position of the statistical region, and the consistency between the deformation direction of the cross-sectional deformation region and the preset deformation direction of the statistical region, the cross-sectional deformation region is classified in the statistical region.

[0118] Based on multiple statistical regions, the cumulative deformation area is statistically analyzed region by region. The deformation depth and deformation width corresponding to the maximum deformation area within the statistical region are obtained, and the deformation ratio is calculated.

[0119] Based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the cross section in the multiple statistical regions, the cross section ruts are classified into structural ruts, unstable ruts, abrasive ruts, or compacted ruts.

[0120] For any region among multiple cross-sectional deformation regions, the average value of the road width direction corresponding to the region is used as the representative abscissa position of the cross-sectional deformation region. Based on the width of the cross-section along the road width direction, the cross-section is divided into multiple statistical regions. In this embodiment, the cross-section is adaptively divided into five non-overlapping statistical regions from the left side of the road to the right side of the road, which are denoted as the first statistical region, the second statistical region, the third statistical region, the fourth statistical region, and the fifth statistical region, respectively.

[0121] Based on the matching of the representative abscissa position of the cross-sectional deformation area with the position of the statistical area, and the consistency between the deformation direction of the cross-sectional deformation area and the preset deformation direction of the statistical area, the cross-sectional deformation area is classified into one or two areas in the statistical area.

[0122] Specifically, the deformation direction of the cross-sectional deformation region is categorized as follows: if the deformation region is below the ideal contour, it is considered positive deformation; if the deformation region is above the ideal contour, it is considered negative deformation. The preset deformation directions for the statistical regions are as follows: the first statistical region has a negative direction; the second statistical region has a positive direction; the third statistical region has a negative direction; the fourth statistical region has a positive direction; and the fifth statistical region has a negative direction. For any region among the multiple cross-sectional deformation regions, a matching statistical region is found based on its representative position. If the deformation direction of the cross-sectional deformation region matches the deformation direction of the matching statistical region, the cross-sectional deformation region is classified as the matching statistical region. If the deformation direction of the cross-sectional deformation region does not match the deformation direction of the matching statistical region, the overlap between the adjacent regions of the matching statistical region and the cross-sectional deformation region is used to determine whether the cross-sectional deformation region belongs to the adjacent region of the matching statistical region. Based on the overlap between adjacent areas of the matching statistical region and the cross-sectional deformation region, it is determined whether the cross-sectional deformation region belongs to the adjacent area of ​​the matching statistical region. Specifically, if the overlap ratio between the adjacent left statistical region of the matching statistical region and the cross-sectional deformation region is greater than a preset percentage threshold, the cross-sectional deformation region is classified as the adjacent left statistical region of the matching statistical region; if the overlap ratio between the adjacent right statistical region of the matching statistical region and the cross-sectional deformation region is greater than a preset percentage threshold, the cross-sectional deformation region is classified as the adjacent right statistical region of the matching statistical region.

[0123] For the five statistical regions mentioned above, the cumulative deformation area SA is calculated for each region, and the deformation depth SR corresponding to the maximum deformation area within each statistical region is obtained. D and deformation width SR W And calculate the deformation ratio. Based on the cumulative deformation area SA and deformation depth SR of the five statistical regions D Deformation width SR W Deformation ratio and the maximum deformation depth D' of the cross section max The cross-sectional ruts are classified into structural ruts, unstable ruts, abrasive ruts, and compacted ruts, thus achieving effective differentiation of rut types.

[0124] In this embodiment, classifying cross-sectional ruts into structural ruts, unstable ruts, abrasive ruts, or compacted ruts based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the cross-section of the multiple statistical regions includes:

[0125] A1: Based on the cumulative deformation area of ​​the multiple statistical regions, the sum of the positive deformation area and the sum of the negative deformation area of ​​the cross section are calculated respectively, and the ratio of the negative deformation area to the positive deformation area is calculated.

[0126] A2: If the ratio of the negative deformation area to the positive deformation area is less than the first preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset first deformation depth threshold, then the cross section rut is classified as a structural rut; otherwise, proceed to step A3.

[0127] A3: If the ratio of the negative deformation area to the positive deformation area is greater than the second preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset second deformation depth threshold, then the cross section rut is classified as an unstable rut; otherwise, proceed to step A4.

[0128] A4: If the deformation ratio of two statistical regions in multiple statistical regions is greater than the preset deformation ratio threshold, and the maximum deformation depth of the cross section is greater than the preset third deformation depth threshold, then the cross section rut is classified as a compacted rut; otherwise, proceed to step A5.

[0129] A5: The cross-sectional ruts are classified as abrasive ruts;

[0130] Wherein, the preset first deformation depth threshold is greater than the preset second deformation depth threshold; the preset second deformation depth threshold is greater than the preset third deformation depth threshold;

[0131] The first preset positive-negative area ratio threshold is less than the second preset positive-negative area ratio threshold.

[0132] Specifically, step A1: Based on the cumulative deformation area SA of the five statistical regions, calculate the sum of the positive deformation area and the sum of the negative deformation area of ​​the cross section, and calculate the ratio R of the negative deformation area to the positive deformation area. NP .

[0133] Step A2: If the ratio R of the negative deformation area to the positive deformation area is... NP The area ratio is less than the first preset positive-negative area ratio threshold (e.g., 0.25), and the maximum deformation depth D' of the cross-section is less than the first preset positive-negative area ratio threshold. max If the depth of deformation exceeds the preset first deformation threshold (e.g., 30mm), the cross-sectional ruts are classified as structural ruts; otherwise, proceed to step A3.

[0134] A3: If the ratio of the negative deformation area to the positive deformation area R NP The area ratio is greater than the second preset positive-negative area ratio threshold (e.g., 0.4), and the maximum deformation depth D' of the cross-section is greater than the second preset positive-negative area ratio threshold. max If the depth of deformation exceeds the preset second deformation threshold (e.g., 15mm), the cross-sectional ruts are classified as unstable ruts; otherwise, proceed to step A4.

[0135] A4: If the deformation ratio R of the second or fourth region DW The deformation ratio is greater than the preset threshold value, and the maximum deformation depth D' of the cross-section is greater than the preset threshold value. max If the depth of deformation exceeds the preset third deformation threshold (e.g., 5mm), the cross-sectional ruts are classified as compacted ruts; otherwise, proceed to step A5.

[0136] A5: Classify cross-sectional ruts as abrasive ruts.

[0137] Through the above steps, based on the ratio of negative deformation area to positive deformation area, combined with the deformation ratio between regions, the rut types can be accurately classified, accurately reflecting the degree of influence of different types of ruts on the pavement structure, thereby providing guidance for road performance evaluation and maintenance plan formulation.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions 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 invention.

Claims

1. A method for classifying ruts based on precise three-dimensional analysis, characterized in that, include: Acquire road surface elevation data and road surface grayscale data; Based on the road surface elevation data and road surface grayscale data, the cross-sectional control profile within the lane range is obtained; Obtain the upper envelope of the cross-sectional control profile, and combine it with the cross-sectional control profile to obtain the ideal cross-sectional control profile. The step of obtaining the upper envelope of the cross-sectional control profile and combining it with the cross-sectional control profile to obtain the ideal cross-sectional control profile includes: Obtain the upper envelope of the cross-sectional control profile; based on the cross-sectional control profile and the upper envelope of the cross-sectional control profile, obtain the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross-section; based on the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross-section, obtain the ideal control profile of the cross-section. The step of obtaining the representative abscissa position and representative elevation value of the unworn areas at both ends of the cross section based on the cross section control profile and its upper envelope includes: The cross-sectional control contour is divided into multiple sub-segments according to a preset step size; for any field, the average absolute distance from all points of the sub-segment in the cross-sectional control contour to the upper envelope is calculated; n sub-segments with an average absolute distance less than a preset distance are selected, and the direction of each sub-segment in the n sub-segments is calculated by linear fitting; based on the direction of each sub-segment in the n sub-segments, the representative direction of the unworn area is obtained; among the multiple sub-segments at the left end of the cross-section, the sub-segment most similar to the representative direction of the sub-segment is selected as the unworn area at the left end, and the unworn area at the left end is assigned to the corresponding sub-segment. The average value of the cross-sectional control profile is used as the representative elevation value of the unworn area at the left end, and the average value of the abscissa of the road width direction corresponding to the unworn area at the left end is used as the representative abscissa position of the unworn area at the left end. Among the multiple sub-segments at the right end of the cross-section, the sub-segment most similar to the representative direction of the sub-segment is selected as the unworn area at the right end, and the average value of the cross-sectional control profile corresponding to the unworn area at the right end is used as the representative elevation value of the unworn area at the right end, and the average value of the road width direction corresponding to the unworn area at the right end is used as the representative abscissa position of the unworn area at the right end. The step of obtaining the ideal control profile of the cross section based on the representative abscissa position and representative elevation value of the unworn areas at the left and right ends of the cross section includes: taking the representative abscissa position and representative elevation value corresponding to the unworn area at the left end as the left endpoint, and taking the representative abscissa position and representative elevation value corresponding to the unworn area at the right end as the right endpoint; connecting the left endpoint and the right endpoint, and taking the connecting line as the ideal control profile of the cross section; Based on the cross-sectional control profile and the ideal cross-sectional control profile, the cross-section is divided into deformation regions to obtain multiple cross-sectional deformation regions. Based on the cross-sectional control profile, the ideal cross-sectional control profile, and multiple cross-sectional deformation regions, the parameters of the cross-sectional deformation regions are obtained. The ruts are classified based on the parameters of the deformation regions of the multiple cross sections.

2. The rut classification method based on precise three-dimensional analysis according to claim 1, characterized in that, The step of obtaining the cross-sectional control profile within the lane range based on the road surface elevation data and road surface grayscale data includes: Based on the road surface elevation data and road surface grayscale data, obtain the target road surface elevation data and target road surface grayscale data within the lane range; Based on any cross section in the target road surface elevation data, obtain the cross section control profile.

3. The rut classification method based on precise three-dimensional analysis according to claim 2, characterized in that, The step of obtaining target road surface elevation data and target road surface grayscale data within the lane range based on the road surface elevation data and road surface grayscale data includes: Based on the road surface elevation data, the first region of potential lane lines is marked using the elevation and geometric features of the road lane lines. Based on the road surface grayscale data, the reflective properties and geometric dimensions of the road lane lines are used to mark the potential second lane line regions; By combining the first potential lane line region and the second potential lane line region, the current lane line position on the road surface is determined; Based on the current lane line position, extract the target road surface elevation data and target road surface grayscale data within the lane range.

4. The rut classification method based on precise three-dimensional analysis according to claim 1, characterized in that, The cross-section is divided into deformation regions based on the cross-sectional control profile and the ideal cross-sectional control profile, resulting in multiple cross-sectional deformation regions, including: For any measuring point in the cross-sectional control profile, calculate the difference between the ideal cross-sectional control profile and the cross-sectional control profile to obtain the set of cross-sectional difference values ​​between the cross-sectional control profile and the ideal cross-sectional control profile. Based on the set of cross-sectional differences, the cross-section is divided into deformation regions according to the continuity of the differences. Regions with continuously positive or continuously negative differences are divided into one region, resulting in multiple cross-sectional deformation regions.

5. The method for rut classification based on precise three-dimensional precision according to any one of claims 1 to 4, characterized in that, The cross-sectional deformation region parameters include: deformation depth, deformation width, and deformation area of ​​the cross-sectional deformation region. The process of obtaining the cross-sectional deformation region parameters based on the cross-sectional control contour, the ideal cross-sectional control contour, and multiple cross-sectional deformation regions includes: For any cross-sectional deformation region, calculate the absolute difference between the cross-sectional control profile and the cross-sectional ideal control profile to obtain the set of absolute differences between the cross-sectional control profile and the cross-sectional ideal control profile deformation regions. For any cross-sectional deformation region, based on the set of absolute differences in deformation regions, the maximum difference in deformation regions is obtained, and the maximum difference in deformation regions is used as the deformation depth of the current cross-sectional deformation region; based on the deformation depths of the multiple cross-sectional deformation regions, the maximum deformation depth of the cross-section is obtained. Calculate the effective deformation depth threshold based on the maximum deformation depth of the cross-section; For any cross-sectional deformation region, count the number of regions whose absolute value is greater than the effective deformation depth threshold in the set of absolute differences of deformation regions corresponding to the cross-sectional deformation region, and obtain the deformation width of the cross-sectional deformation region by combining the lateral sampling interval of the precise three-dimensional point cloud data. For any cross-sectional deformation region, the deformation area of ​​the cross-sectional deformation region is calculated based on the deformation depth and deformation width of the cross-sectional deformation region.

6. The rut classification method based on precise three-dimensional analysis according to claim 5, characterized in that, The classification of ruts based on the parameters of the multiple cross-sectional deformation regions includes: For any region among the multiple cross-sectional deformation regions, the average value of the road width direction corresponding to the region is taken as the representative abscissa position of the cross-sectional deformation region. Based on the width of the cross section along the road width direction, the cross section is divided into multiple statistical regions; For any region among the multiple cross-sectional deformation regions, based on the matching of the representative abscissa position of the cross-sectional deformation region with the position of the statistical region, and the consistency between the deformation direction of the cross-sectional deformation region and the preset deformation direction of the statistical region, the cross-sectional deformation region is classified in the statistical region. Based on multiple statistical regions, the cumulative deformation area is statistically analyzed region by region. The deformation depth and deformation width corresponding to the maximum deformation area within the statistical region are obtained, and the deformation ratio is calculated. Based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the cross section in the multiple statistical regions, the cross section ruts are classified into structural ruts, unstable ruts, abrasive ruts, or compacted ruts.

7. The rut classification method based on precise three-dimensional analysis according to claim 6, characterized in that, The classification of cross-sectional ruts into structural ruts, unstable ruts, abrasive ruts, or compacted ruts based on the cumulative deformation area, deformation depth, deformation width, deformation ratio, and maximum deformation depth of the multiple statistical regions includes: A1: Based on the cumulative deformation area of ​​the multiple statistical regions, the sum of the positive deformation area and the sum of the negative deformation area of ​​the cross section are calculated respectively, and the ratio of the negative deformation area to the positive deformation area is calculated. A2: If the ratio of the negative deformation area to the positive deformation area is less than the first preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset first deformation depth threshold, then the cross section rut is classified as a structural rut; otherwise, proceed to step A3. A3: If the ratio of the negative deformation area to the positive deformation area is greater than the second preset positive-negative area ratio threshold, and the maximum deformation depth of the cross section is greater than the preset second deformation depth threshold, then the cross section rut is classified as an unstable rut; otherwise, proceed to step A4. A4: If the deformation ratio of two statistical regions in multiple statistical regions is greater than the preset deformation ratio threshold, and the maximum deformation depth of the cross section is greater than the preset third deformation depth threshold, then the cross section rut is classified as a compacted rut; otherwise, proceed to step A5. A5: The cross-sectional ruts are classified as abrasive ruts; Wherein, the preset first deformation depth threshold is greater than the preset second deformation depth threshold; the preset second deformation depth threshold is greater than the preset third deformation depth threshold; The first preset positive-negative area ratio threshold is less than the second preset positive-negative area ratio threshold.

Images