A disease three-dimensional information extraction method based on a precise three-dimensional pavement

Abstract

The application provides a disease three-dimensional information extraction method based on a precise three-dimensional pavement, and relates to the field of disease three-dimensional information extraction, and comprises the following steps: acquiring the representative length, the representative width, the disease transverse expansion range and the disease longitudinal expansion range of each disease type according to the precise three-dimensional pavement data and the disease type information; determining the disease influence length and the disease influence width corresponding to each disease type according to the representative length, the representative width, the disease transverse expansion range and the disease longitudinal expansion range of each disease type; and extracting the three-dimensional information of the pavement disease according to the disease influence length, the disease influence width and the disease influence depth. According to the disease influence length, the disease influence width and the disease influence depth, the three-dimensional information of the pavement disease is extracted, the disease three-dimensional information which can accurately represent the disease is extracted from the three-dimensional pavement, the pavement damage is accurately evaluated, the reliability of the three-dimensional information extraction is greatly improved, and a good foundation is provided for the development of three-dimensional detection technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of three-dimensional information extraction of diseases, and in particular to a three-dimensional information extraction method of diseases based on a precise three-dimensional road surface. BACKGROUND

[0002] In the process of three-dimensional information extraction of road surface diseases, different disease areas have different disease characteristics at different disease locations, which makes the extracted three-dimensional information of diseases unable to accurately reflect and evaluate the real situation of road damage. SUMMARY

[0003] The present application provides a three-dimensional information extraction method of diseases based on a precise three-dimensional road surface, which solves the technical defect that the current road damage situation cannot be accurately evaluated for different disease types. The present application proposes a method of using the elevation difference, surface curvature difference or cross-sectional curve slope difference characteristics of the disease area and non-disease area near the disease boundary area, combining with the disease representative three-dimensional information, estimating the expansion influence range of the disease, and then realizing the accurate evaluation of the road damage situation.

[0004] In a first aspect, the present application provides a three-dimensional road surface disease extraction method, comprising:

[0005] According to the precise three-dimensional road surface data and the disease type information, the representative length, the representative width, the disease lateral expansion range and the disease longitudinal expansion range of each disease type are obtained; and the disease influence length and the disease influence width corresponding to each disease type are determined according to the representative length, the representative width, the disease lateral expansion range and the disease longitudinal expansion range of each disease type.

[0006] According to the disease influence length, the disease influence width and the disease influence depth, the three-dimensional information of the road surface disease is extracted.

[0007] In the case of linear disease, the representative length is determined according to the diagonal line length of the circumscribed rectangle of the linear disease, the representative width is determined according to the average width of the linear disease and the representative maximum width of the linear disease, the disease lateral expansion range is a first preset constant, and the disease longitudinal expansion range is a second preset constant.

[0008] In the case of surface disease, the representative length is determined according to the representative maximum length of the surface disease, the representative width is determined according to the representative maximum width of the surface disease, the disease lateral expansion range is a lateral expansion disease point selected from a first preset area according to a lateral segmentation threshold, and the disease longitudinal expansion range is a longitudinal expansion disease point selected from the first preset area according to a longitudinal segmentation threshold.

[0009] The first preset area is an area that is non-overlapping with the disease area in a first preset radius range and is centered on a disease area positioning point.

[0010] The disease three-dimensional information extraction method based on the precise three-dimensional pavement provided by the application further comprises the following steps before determining the disease influence length and the disease influence width corresponding to each disease type:

[0011] After the second preset area and the disease area are extracted from the total pavement area, an ideal pavement area is obtained to filter a reference pavement according to the ideal pavement area.

[0012] The second preset area is an area that is non-overlapping with the disease area in a second preset radius range and is centered on a disease area positioning point.

[0013] The disease three-dimensional information extraction method based on the precise three-dimensional pavement provided by the application, wherein the representative length is determined according to the diagonal length of the circumscribed rectangle of the linear disease, and the method comprises the following steps:

[0014] The length and the width of the circumscribed rectangle of the linear disease area are obtained.

[0015] The diagonal length of the circumscribed rectangle is calculated according to the length and the width of the circumscribed rectangle to obtain the representative length of the linear disease.

[0016] The disease three-dimensional information extraction method based on the precise three-dimensional pavement provided by the application, wherein the representative width is determined according to the average width of the linear disease and the representative maximum width of the linear disease, and the method comprises the following steps:

[0017] For any disease point, a representative direction of a set of all points in a third preset area centered on the disease point is obtained.

[0018] If the representative direction and the transverse included angle are less than or equal to a preset angle, the number of longitudinally continuous points at the position of the disease point is counted to obtain the disease point width according to the number of longitudinally continuous points and a longitudinal sampling interval; otherwise, the number of transversely continuous points at the position of the disease point is counted to obtain the disease point width according to the number of transversely continuous points and a transverse sampling interval.

[0019] All disease point widths are obtained by traversing all disease points to determine the average width of the linear disease according to the disease point width of each disease point.

[0020] The number of width data is determined according to the number of disease points of the disease area, and the initial representative maximum width determined according to the number of width data and a preset representative maximum width quantile is taken as the representative maximum width of the linear disease; or, the mean of the width data greater than the initial representative maximum width in the width distribution is taken as the representative maximum width of the linear disease.

[0021] determining a first weighted value according to the average width of linear diseases and a first weighted coefficient;

[0022] determining a second weighted value according to the representative maximum width of linear diseases and a second weighted coefficient;

[0023] determining the representative width according to the first weighted value and the second weighted value;

[0024] a sum of the first weighted coefficient and the second weighted coefficient is a third preset constant.

[0025] According to the application, the representative length is determined according to the representative maximum length of surface diseases, and the method comprises the following steps.

[0026] determining the number of length data according to the position information of the surface disease area, and taking an initial representative maximum length determined according to the number of length data and a preset representative maximum length quantile as the representative maximum length of surface diseases;

[0027] or, taking the average of length data greater than the initial representative maximum length in the length distribution as the representative maximum length of surface diseases.

[0028] According to the application, the representative width is determined according to the representative maximum width of surface diseases, and the method comprises the following steps.

[0029] determining the number of width data according to the position information of the surface disease area, and taking an initial representative maximum width determined according to the number of width data and a preset representative maximum width quantile as the representative maximum width of surface diseases;

[0030] or, taking the average of width data greater than the initial representative maximum width in the width distribution as the representative maximum width of surface diseases.

[0031] According to the application, the lateral expansion range of diseases is determined according to the lateral expansion disease points screened out from the first preset area according to the lateral segmentation threshold, and the method comprises the following steps.

[0032] determining the disease depth distribution and the number of depth data according to the disease point depth of all disease points in the surface disease area, and taking an initial representative maximum depth determined according to the number of depth data and a preset representative maximum depth quantile as the representative maximum depth of surface diseases; or, taking the average of depth data greater than the initial representative maximum depth in the depth distribution as the representative maximum depth of surface diseases;

[0033] determining a fourth preset area where the disease area is located, determining a height difference of a height of any point in the fourth preset area and a height of a corresponding position of the reference road surface, obtaining an average of all height differences after traversing all points in the fourth preset area, and determining a horizontal segmentation threshold according to the average of all height differences and a maximum depth of the planar disease;

[0034] traversing all points in the second preset area, and determining a point where a height difference of a height of any point and a height of a corresponding position of the reference road surface is greater than the horizontal segmentation threshold as a horizontally expanded disease point;

[0035] after denoising all horizontally expanded disease points, determining an average range and a maximum range of horizontal disease expansion, and determining a horizontal disease expansion range according to the average range and the maximum range of the horizontal disease expansion;

[0036] the fourth preset area is a region in cross-sectional data where the disease area is located, and the disease area and the second preset area are removed.

[0037] According to the disease longitudinal expansion range determined by the longitudinal segmentation threshold from the first preset area, the disease longitudinal expansion range is determined by the longitudinal segmentation threshold from the first preset area, including:

[0038] determining a disease depth distribution and a number of depth data according to a disease point depth of all disease points in the planar disease area, taking an initial maximum representative depth determined according to the number of depth data and a preset maximum representative depth quantile as the maximum representative depth of the planar disease, or taking a depth data mean of depth data greater than the initial maximum representative depth in the depth distribution as the maximum representative depth of the planar disease;

[0039] determining a fifth preset area where the disease area is located, determining a height difference of a height of any point in the fifth preset area and a height of a corresponding position of the reference road surface, obtaining an average of all height differences after traversing all points in the fifth preset area, and determining a longitudinal segmentation threshold according to the average of all height differences and a maximum depth of the planar disease;

[0040] traversing all points in the second preset area, and determining a point where a height difference of a height of any point and a height of a corresponding position of the reference road surface is greater than the longitudinal segmentation threshold as a longitudinally expanded disease point;

[0041] after denoising all longitudinally expanded disease points, determining an average range and a maximum range of longitudinal disease expansion, and determining a longitudinal disease expansion range according to the average range and the maximum range of the longitudinal disease expansion;

[0042] the fifth preset area is a region in longitudinal cross-sectional data where the disease area is located, and the disease area and the second preset area are removed.

[0043] The method for extracting three-dimensional information of diseases based on a precise three-dimensional road surface according to the present application extracts three-dimensional information of road diseases according to disease influence length, disease influence width and disease influence depth, and comprises:

[0044] In the case of linear disease, for any disease point in the disease area, the difference between the elevation of the disease point and the elevation of the corresponding position of the reference road surface is taken as the disease point depth, all disease points in the disease area are traversed, the disease depths of all disease points in the disease area are obtained, and the disease depths of all disease points are averaged to obtain the average disease depth of the linear disease.

[0045] The disease depth distribution and the number of depth data are determined according to the disease point depths of all disease points in the disease area of the linear disease, the initial representative maximum depth is determined according to the number of depth data and a preset representative maximum depth quantile, and the initial representative maximum depth is taken as the disease representative maximum depth of the linear disease; or, the average of the depth data greater than the initial representative maximum depth in the depth distribution is taken as the disease representative maximum depth of the linear disease.

[0046] The third weighting value is determined according to the average disease depth of the linear disease and a third weighting coefficient.

[0047] The fourth weighting value is determined according to the disease representative maximum depth of the linear disease and a fourth weighting coefficient.

[0048] The influence depth of the linear disease is determined according to the third weighting value and the fourth weighting value.

[0049] The three-dimensional information of the linear disease in the three-dimensional road surface is extracted according to the disease influence length, the disease influence width and the influence depth of the linear disease.

[0050] In the case of surface disease, the surface disease influence depth is obtained according to the surface disease representative maximum depth and a preset surface disease minimum depth.

[0051] The three-dimensional information of the surface disease in the three-dimensional road surface is extracted according to the disease influence length, the disease influence width and the surface disease influence depth.

[0052] The method for extracting three-dimensional information of diseases based on a precise three-dimensional road surface according to the present application further comprises, before determining the disease influence length and the disease influence width corresponding to each disease type according to the representative length, the representative width, the disease transverse expansion range and the disease longitudinal expansion range of each disease type:

[0053] The disease ratio is determined according to the circumscribed rectangle length, the width of the disease area and the area of the disease area.

[0054] In a case that the disease ratio is greater than or equal to a preset ratio and the circumscribed rectangle width is greater than a preset width, the disease type is determined as a planar disease;

[0055] In a case that the disease ratio is less than the preset ratio or the circumscribed rectangle width is less than or equal to the preset width, the disease type is determined as a linear disease.

[0056] The planar disease includes a macroscopic deformation disease and / or a reticular disease.

[0057] The application provides a disease three-dimensional information extraction method based on a precise three-dimensional pavement, according to the precise three-dimensional pavement data and the disease type information, the representative length, the representative width, the disease lateral extension range and the disease longitudinal extension range of each disease type are obtained, the disease influence length and the disease influence width corresponding to different disease types are determined according to the representative length, the representative width, the disease lateral extension range and the disease longitudinal extension range of different disease types, and then the three-dimensional information of the pavement disease is extracted according to the disease influence length, the disease influence width and the disease influence depth, the disease three-dimensional information which can accurately characterize the disease can be extracted from the three-dimensional pavement, the pavement damage is accurately evaluated, the reliability of the disease three-dimensional information extraction is greatly improved, and a good foundation is provided for the development of the three-dimensional detection technology. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0059] Figure 1 Fig. 1 is a flowchart of the disease three-dimensional information extraction method based on the precise three-dimensional pavement provided by the application;

[0060] Figure 2 Fig. 2 is a flowchart of determining the representative length of the linear disease provided by the application;

[0061] Figure 3 Fig. 3 is a flowchart of determining the representative width of the linear disease provided by the application;

[0062] Figure 4 Fig. 4 is a flowchart of determining the disease lateral extension range of the planar disease provided by the application;

[0063] Figure 5 Fig. 5 is a flowchart of determining the disease longitudinal extension range of the planar disease provided by the application;

[0064] Figure 6This is a schematic diagram of the process for extracting three-dimensional information of road surface defects provided by the present invention;

[0065] Figure 7 This is a flowchart illustrating the method for obtaining the types of road defects based on precise three-dimensional pavement, as per the present invention. Detailed Implementation

[0066] 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.

[0067] The performance of highways directly impacts urban development. During their service life, under the combined effects of traffic loads and natural environmental conditions, highway surfaces gradually develop damage such as cracks, ruts, potholes, bulges, subsidence, and misalignment, leading to a gradual decline in performance. This affects vehicle speed and driving safety to varying degrees, significantly impacting overall economic and social benefits. According to relevant regulations, the technical condition of asphalt pavements must be regularly inspected and evaluated, with timely updates to pavement condition data and analysis of the causes of pavement damage. In the detection of surface-level pavement damage, traditional damage detection methods using two-dimensional image detection technology cannot obtain depth information about pavement damage. Without depth information, accurate evaluation of pavement damage is impossible.

[0068] This invention employs a three-dimensional pavement inspection method, integrating numerous modern advanced information technologies to elevate pavement inspection data from two-dimensional to three-dimensional. This significantly enriches pavement condition data, enabling automated detection of various types of defects and greatly improving the efficiency and reliability of highway inspection. The three-dimensional inspection technology used in this invention provides a solid foundation for accurate assessment of pavement damage. It accurately extracts three-dimensional information about pavement defects from the three-dimensional pavement, providing a necessary prerequisite for precise evaluation of pavement damage based on this information. For any given defect, different locations exhibit different characteristics; this invention can extract accurate three-dimensional information about the defect from the three-dimensional pavement.

[0069] Figure 1 This is a flowchart illustrating the method for extracting three-dimensional information on road surface defects based on a precise three-dimensional pavement, provided by the present invention. The present invention provides a method for extracting three-dimensional information on road surface defects based on a precise three-dimensional pavement, comprising:

[0070] Based on precise three-dimensional pavement data and disease type information, obtain the representative length, representative width, lateral expansion range, and longitudinal expansion range of each disease type.

[0071] The length and width of each disease type are determined based on its representative length, representative width, horizontal expansion range, and vertical expansion range.

[0072] Three-dimensional information of pavement defects is extracted based on the length, width, and depth of the defects.

[0073] When the disease type is linear disease, the representative length is determined based on the length of the diagonal of the circumscribed moment of the linear disease, the representative width is determined based on the average width of the linear disease and the maximum representative width of the linear disease, the lateral expansion range of the disease is a first preset constant, and the longitudinal expansion range of the disease is a second preset constant.

[0074] When the disease type is a planar disease, the representative length is determined based on the maximum representative length of the planar disease, the representative width is determined based on the maximum representative width of the planar disease, the lateral expansion range of the disease is the lateral expansion disease points selected from the first preset region based on the lateral segmentation threshold, and the longitudinal expansion range of the disease is the longitudinal expansion disease points selected from the first preset region based on the longitudinal segmentation threshold.

[0075] The first preset area is the area that does not overlap with the diseased area within a first preset radius, centered on the location point of the diseased area.

[0076] In step 101, based on the precise three-dimensional pavement data and the disease type information, the representative length, representative width, lateral expansion range, and longitudinal expansion range of each disease type are obtained; the precise three-dimensional pavement data is obtained through a precise three-dimensional measurement sensor.

[0077] Optionally, the precision three-dimensional measurement sensor may be a line-scanning three-dimensional measurement sensor, which includes a laser and a high-speed 3D camera. The laser is used to project a laser beam perpendicularly onto the road surface, and the high-speed 3D camera is at a certain angle to the laser to acquire cross-sectional data corresponding to the location of the laser line. A controller is used to control the sensor to acquire cross-sectional data of the road surface. The cross-sectional data includes the elevation and grayscale of the road surface corresponding to the laser line. The cross-sectional data is assembled into three-dimensional contour data of the road surface along the acquisition time sequence. Three-dimensional modeling is performed on the three-dimensional contour data of the road surface to obtain the precision three-dimensional road surface data.

[0078] In step 102, the disease influence length and disease influence width corresponding to each disease type are determined based on the representative length, representative width, lateral expansion range, and longitudinal expansion range of each disease type. The disease types include planar diseases and linear diseases, and the linear diseases include at least linear diseases and network diseases. The calculation methods for obtaining the representative length, representative width, lateral expansion range, and longitudinal expansion range are also different for different disease types.

[0079] For linear diseases, the representative length is determined based on the length of the diagonal of the circumscribed rectangle of the linear disease, and the representative width is determined based on the average width of the linear disease and the maximum representative width of the linear disease. The lateral expansion range of the disease is a first preset constant, and the longitudinal expansion range of the disease is a second preset constant. For area diseases, the representative length is determined based on the maximum representative length of the area disease, and the representative width is determined based on the maximum representative width of the area disease. The lateral expansion range of the disease is the lateral expansion disease points selected from the first preset region based on a lateral segmentation threshold, and the longitudinal expansion range of the disease is the longitudinal expansion disease points selected from the first preset region based on a longitudinal segmentation threshold.

[0080] In step 103, by combining different types of defects, the corresponding defect length, defect width, and defect depth can be determined, and the three-dimensional information of the road defects can be extracted based on the defect length, defect width, and defect depth.

[0081] Optionally, before determining the length, width, and depth of disease impact corresponding to each disease type, the method further includes:

[0082] After extracting the second preset area and the defect area from the total road surface area, the ideal road surface area is obtained, and the reference road surface is filtered out based on the ideal road surface area.

[0083] The second preset area is the area that does not overlap with the diseased area within a second preset radius, centered on the location point of the diseased area.

[0084] This invention divides non-disease areas into suspected disease areas and normal road surface areas. The suspected disease area is defined by expanding the disease boundary with a preset radius of influence, centered on the disease area location result. The expanded area is designated as the suspected disease area, also known as the second preset area. By extracting the second preset area and the disease area from the total road surface area, the direct or indirect influence of the disease area and the suspected disease area on the ideal road surface area can be completely eliminated. For the ideal road surface area, a filtering method is used to obtain the corresponding reference surface, and the reference surface is determined as the reference road surface.

[0085] More specifically, in the total road surface area, after removing the normal road surface area, abnormal road surface areas are determined. The abnormal road surface areas include suspected defect areas and defect areas. The method used to determine the reference road surface is the same as that used for the ideal road surface area. Based on the initial result obtained from the reference surface corresponding to the normal road surface area, the present invention can obtain the reference surface corresponding to the abnormal road surface area by extending the reference surface.

[0086] In an optional embodiment, the present invention can first use cross-sections as a basis, and use the corresponding points in the reference planes of the left and right normal road areas adjacent to the abnormal road area in the cross-section data as extension endpoints to supplement abnormal road data in a linear manner section by section; for the remaining unsupplemented abnormal road data, use longitudinal sections as a basis, and use the corresponding points in the reference planes of the upper and lower normal road areas adjacent to the abnormal road area in the longitudinal section data as extension endpoints to supplement abnormal road data in a linear manner section by section; for abnormal road data that is still not supplemented after cross-section supplementation and longitudinal section supplementation, the abnormal road data is supplemented by a triangular network construction method.

[0087] In another optional embodiment, the present invention can first use the longitudinal profile as a basis, and then use the corresponding points in the reference planes of the upper and lower normal road areas adjacent to the abnormal road area in the longitudinal profile data as extension endpoints to supplement the abnormal road surface data in a linear manner section by section; for the remaining abnormal road surface data that has not been supplemented, the cross section is used as a basis, and the corresponding points in the reference planes of the left and right normal road areas adjacent to the abnormal road area in the cross section data are used as extension endpoints to supplement the abnormal road surface data in a linear manner section by section; for the abnormal road surface data that has not been supplemented after longitudinal profile supplementation and cross section supplementation, the abnormal road surface data is supplemented by using a triangular network construction method.

[0088] Optionally, the representative length is determined based on the maximum representative length of the areal disease, including:

[0089] The number of length data points is determined based on the location information of the area of ​​the disease, and the initial representative maximum length, determined based on the number of length data points and the preset representative maximum length quantile, is taken as the representative maximum length of the disease.

[0090] Alternatively, the average length of the length data in the length distribution that is greater than the initial representative maximum length can be used as the representative maximum length of the areal disease.

[0091] In an optional embodiment, the present invention can calculate the representative maximum length according to a preset representative maximum length quantile. Specifically, the length distribution data is sorted to obtain a length dataset arranged in ascending or descending order. Based on the length dataset arranged in ascending or descending order, the initial representative maximum length is determined by the product of the number of length data and the preset representative maximum length quantile, and the initial representative maximum length is determined as the representative maximum length of the areal disease.

[0092] In another alternative embodiment, the initial representative maximum length is compared with all length data in the length distribution, and length data greater than the initial representative maximum length are determined. All length data greater than the initial representative maximum length are averaged, and the result after averaged processing is determined as the representative maximum length of the areal disease.

[0093] Optionally, the representative width is determined based on the maximum representative width of the areal disease, including:

[0094] The number of width data points is determined based on the location information of the area of ​​the disease, and the initial representative maximum width, determined based on the number of width data points and the preset representative maximum width quantile, is used as the representative maximum width of the disease.

[0095] Alternatively, the average width of the width data in the width distribution that is greater than the initial representative maximum width can be used as the representative maximum width of the areal disease.

[0096] In an optional embodiment, the present invention can calculate the representative maximum width according to a preset representative maximum width quantile. Specifically, the width distribution data is sorted to obtain a width dataset arranged in ascending or descending order. Based on the width dataset arranged in ascending or descending order, the initial representative maximum width is determined by the product of the number of width data and the preset representative maximum width quantile, and the initial representative maximum width is determined as the representative maximum width of the areal disease.

[0097] In another alternative embodiment, the initial representative maximum width is compared with all width data in the width distribution, and width data that are greater than the initial representative maximum width are determined. All width data that are greater than the initial representative maximum width are averaged, and the result after averaged processing is determined as the representative maximum width of the areal disease.

[0098] This invention provides a method for extracting three-dimensional information of pavement defects based on precise three-dimensional pavement data. Based on precise three-dimensional pavement data and defect type information, the method obtains the representative length, representative width, lateral extension range, and longitudinal extension range of each defect type. Based on these parameters, the method determines the corresponding defect impact length and width for each defect type. Then, it extracts the three-dimensional information of pavement defects based on the defect impact length, width, and depth. This invention can extract accurate three-dimensional defect information from three-dimensional pavement and provide precise evaluation of pavement damage, greatly improving the reliability of three-dimensional defect information extraction and providing a solid foundation for the development of three-dimensional detection technology.

[0099] Figure 2 This is a flowchart illustrating the process for determining the representative width of linear diseases provided by the present invention. The representative length is determined based on the width of the diagonal of the circumscribed moment of the linear disease, including:

[0100] Obtain the length and width of the circumscribed radius of the linear disease area;

[0101] Calculate the diagonal width of the circumscribed moment based on its length and width to obtain the representative length of the linear disease.

[0102] In step 201, the three-dimensional pavement in this invention does not only include one type of defect, but may face a complex situation where multiple defect types are combined. In such cases, this invention will divide the defect area into planar defect area and linear defect area, and provide different calculation methods for different defect types. For the representative width of each linear defect area, the length and width of the circumscribed moment of the defect area must be obtained first.

[0103] In step 202, after determining the length and width of the circumscribed moment of the linear disease area, the width of the diagonal of the circumscribed moment is calculated using the Pythagorean theorem based on the length and width of the circumscribed moment, and the length of the diagonal of the circumscribed moment is determined as the representative length of the linear disease.

[0104] This invention provides a completely different method for calculating the representative length of a disease based on the characteristics of different disease types. The representative width is used as a reference for the extraction of three-dimensional information of the disease, which is described later. This improves the calculation accuracy and the accuracy of extracting three-dimensional information of the disease.

[0105] Figure 3 This is a flowchart illustrating the process for determining the representative width of linear diseases provided by the present invention. The representative width is determined based on the average width of the linear disease and the maximum representative width of the linear disease, including:

[0106] For any disease point, take the disease point as the center and obtain the representative direction of the set of all points in the third preset area;

[0107] If the angle between the representative direction and the lateral direction is less than or equal to a preset angle, count the number of consecutive points in the longitudinal direction at the location of the disease point, and obtain the width of the disease point based on the number of consecutive points in the longitudinal direction and the longitudinal sampling interval; otherwise, count the number of consecutive points in the lateral direction at the location of the disease point, and obtain the width of the disease point based on the number of consecutive points in the lateral direction and the lateral sampling interval.

[0108] Traverse all disease points and obtain the width of each disease point to determine the average width of the linear disease based on the width of each disease point.

[0109] The number of width data points is determined based on the number of disease points in the diseased area. The initial representative maximum width, determined based on the number of width data points and the preset representative maximum width quantile, is taken as the representative maximum width of the linear disease; or, the average width data in the width distribution that is greater than the initial representative maximum width is taken as the representative maximum width of the linear disease.

[0110] The first weighting value is determined based on the average width of the linear disease and the first weighting coefficient.

[0111] The second weighting value is determined based on the maximum width represented by linear diseases and the second weighting coefficient;

[0112] The representative width is determined based on the first weighted value and the second weighted value;

[0113] The sum of the first weighting coefficient and the second weighting coefficient is a third preset constant.

[0114] In step 301, for any disease point, taking the disease point as the center, obtain the representative direction of the set of all points in the third preset area, where the third preset area is a preset area determined with the disease point as the center and a preset length as the radius.

[0115] In step 302, in an optional embodiment, for the width of any point in the linear disease area, the effective area is determined with the disease point as the center and a preset length as the radius. That is, the angle between the current disease and the horizontal or vertical direction is calculated for the disease points in the third preset area. The preset angle can be 45°. That is, if the angle with the horizontal direction is less than or equal to 45°, the width of the current point is obtained by counting the number of consecutive points in the vertical direction and multiplying the number of consecutive points in the vertical direction by the vertical sampling interval; otherwise, the width of the current point is obtained by counting the number of consecutive points in the horizontal direction and multiplying the number of consecutive points in the horizontal direction by the horizontal sampling interval.

[0116] In step 303, all disease points are traversed to obtain the width of all disease points, so as to determine the average width of the linear disease based on the width of each disease point. Optionally, the present invention sorts the width of all points to obtain the width distribution and calculates the mean of the width of all points to obtain the average width.

[0117] In step 304, the number of width data points is determined based on the number of disease points in the diseased area, and the initial representative maximum width determined based on the number of width data points and the preset representative maximum width quantile is taken as the representative maximum width of the linear disease; or, the average width data in the width distribution that is greater than the initial representative maximum width is taken as the representative maximum width of the linear disease.

[0118] In one optional embodiment, the method for determining the representative maximum width of a planar disease in this invention can refer to the method for determining the representative maximum width of a planar disease. The representative maximum width can be calculated according to a preset representative maximum width quantile. Specifically, the number of data points in the width distribution data and the preset representative maximum width quantile are determined. The product of the number of width data points and the preset representative maximum width quantile is used to determine the initial representative maximum width, and this initial representative maximum width is determined as the representative maximum width of the linear disease. In another optional embodiment, the initial representative maximum width is compared with all width data points in the width distribution, and width data points greater than the initial representative maximum width are determined. All width data points greater than the initial representative maximum width are averaged, and the result of the averaged processing is determined as the representative maximum width of the linear disease.

[0119] In step 305, a first weighted value is determined based on the product of the average width of the linear disease and a first weighting coefficient, wherein the first weighting coefficient is a preset weighting coefficient.

[0120] In step 306, a second weighting value is determined based on the product of the maximum width represented by the linear disease and the second weighting coefficient. The first weighting coefficient and the second weighting coefficient satisfy certain conditions, and the sum of the first weighting coefficient and the second weighting coefficient is a third preset constant. The third preset constant can be 1. Optionally, the first weighting coefficient is 0.3 and the second weighting coefficient is 0.7; or the first weighting coefficient is 0.45 and the second weighting coefficient is 0.55.

[0121] In step 307, the representative width is determined based on the first weighted value and the second weighted value. For example, if the linear disease is a linear disease, the representative width can be determined by referring to the following formula:

[0122] WR=W W1 *WA+W W2 *WM (1)

[0123] Where WR represents the width of the linear disease, W W1 W is the first weighting coefficient. W2 WA is the average width of the linear disease, and WM is the maximum representative width of the linear disease.

[0124] Figure 4 This is a flowchart illustrating the process for determining the lateral expansion range of planar diseases provided by the present invention. The lateral expansion range is determined from the laterally expanding disease points selected from a first preset region based on a lateral segmentation threshold, including:

[0125] The depth distribution and number of depth data points are determined based on the depth of all disease points within the area of ​​the disease. The initial representative maximum depth, determined based on the number of depth data points and a preset representative maximum depth quantile, is taken as the representative maximum depth of the disease. Alternatively, the average depth data points in the depth distribution that are greater than the initial representative maximum depth are taken as the representative maximum depth of the disease.

[0126] The fourth preset area where the disease area is located is determined. The elevation difference between any point in the fourth preset area and the corresponding position of the reference road surface is obtained by traversing all points in the fourth preset area. The lateral segmentation threshold is determined based on the average of all elevation differences and the maximum representative depth of the surface disease.

[0127] Traverse all points in the second preset area, and determine the points whose elevation difference between any point and the corresponding position of the reference road surface is greater than the horizontal segmentation threshold as horizontal expansion disease points;

[0128] After denoising all laterally expanding disease points, determine the average and maximum range of lateral disease expansion, and then determine the lateral expansion range of the disease based on the average and maximum range of lateral disease expansion.

[0129] The fourth preset region is the region in the cross-sectional data where the diseased region is located, after removing the diseased region and the second preset region.

[0130] In step 401, the representative maximum depth can be calculated according to a preset representative maximum depth quantile. Specifically, the depth distribution data is sorted to obtain a depth dataset arranged in ascending or descending order. Based on the depth dataset arranged in ascending or descending order, the initial representative maximum depth is determined by multiplying the number of depth data points by the preset representative maximum depth quantile, and this initial representative maximum depth is determined as the representative maximum depth of the linear disease. In another optional embodiment, the initial representative maximum depth is compared with all depth data in the depth distribution, and depth data points greater than the initial representative maximum depth are identified. All depth data points greater than the initial representative maximum depth are averaged, and the averaged result is determined as the representative maximum depth of the linear disease.

[0131] In step 402, the fourth preset region is the region in the cross-sectional data where the diseased region is located, after removing the diseased region and the second preset region. The fourth preset region is a potential diseased region. The average elevation difference between any point in the fourth preset region and the corresponding position elevation of the reference road surface is calculated for each cross-section. Combined with the maximum depth represented by linear diseases, the cross-sectional elevation difference segmentation threshold is obtained through a weighted average algorithm or mean-based processing.

[0132] In step 403, all points in the second preset area are traversed, and points whose elevation difference with the corresponding position of the reference road surface is greater than the lateral segmentation threshold are identified as lateral expansion disease points. Based on the cross-sectional elevation difference segmentation threshold, this invention discusses whether each point in the potential fourth preset area is an expansion disease point on a cross-section basis. Specifically, points whose elevation difference is greater than the lateral segmentation threshold are marked as lateral expansion disease points, and the lateral expansion disease points are recorded in the expansion binary map.

[0133] In step 404, after denoising all laterally expanding disease points, the average and maximum ranges of the laterally expanding disease are determined. Based on the average and maximum ranges of the laterally expanding disease, the lateral expansion range of the disease is determined. This invention utilizes a morphological processing method, employing a dilation-then-erosion approach to extend the disease. Based on the length characteristics of connected regions, shorter noise regions are removed, thereby determining the denoised laterally expanding disease points. The average and maximum ranges of the laterally expanding disease are statistically analyzed, and the influence range of the lateral expansion of the disease is calculated. This invention can use a weighted algorithm to process the average and maximum ranges of the lateral expansion of the disease, thereby determining the influence range of the lateral expansion of the disease.

[0134] Figure 5 This is a flowchart illustrating the process for determining the longitudinal expansion range of planar diseases provided by the present invention. The longitudinal expansion range of the disease is determined from the longitudinally expanding disease points selected from a first preset region based on a longitudinal segmentation threshold, including:

[0135] The depth distribution and number of depth data points are determined based on the depth of all disease points within the area of ​​the disease. The initial representative maximum depth, determined based on the number of depth data points and a preset representative maximum depth quantile, is taken as the representative maximum depth of the disease. Alternatively, the average depth data points in the depth distribution that are greater than the initial representative maximum depth are taken as the representative maximum depth of the disease.

[0136] The fifth preset area where the disease area is located is determined. For any point in the fifth preset area, the elevation difference between the elevation of the point and the corresponding position of the reference road surface is obtained by traversing all points in the fifth preset area. The longitudinal segmentation threshold is determined based on the average of all elevation differences and the maximum representative depth of the surface disease.

[0137] Traverse all points in the second preset area, and determine the points whose elevation difference between any point and the corresponding position of the reference road surface is greater than the longitudinal segmentation threshold as longitudinal expansion disease points;

[0138] After denoising all longitudinally expanding disease points, determine the average and maximum range of longitudinal disease expansion, and then determine the longitudinal expansion range of the disease based on the average and maximum range of the longitudinal disease expansion.

[0139] The fifth preset region is the region in the longitudinal section data where the diseased region is located, after removing the diseased region and the second preset region.

[0140] In step 501, referring to step 401, the representative maximum depth can be calculated according to a preset representative maximum depth quantile. Specifically, the depth distribution data is sorted to obtain a depth dataset arranged in ascending or descending order. Based on the depth dataset arranged in ascending or descending order, the initial representative maximum depth is determined by multiplying the number of depth data points by the preset representative maximum depth quantile, and this initial representative maximum depth is determined as the representative maximum depth of the areal disease. In another optional embodiment, the initial representative maximum depth is compared with all depth data in the depth distribution, and depth data points greater than the initial representative maximum depth are identified. All depth data points greater than the initial representative maximum depth are averaged, and the averaged result is determined as the representative maximum depth of the areal disease.

[0141] In step 502, the fifth preset region is the region in the longitudinal profile data where the defect area is located, after removing the defect area and the second preset region. The fifth preset region is a potential defect area. The average elevation difference between any point in the fourth preset region and the corresponding position elevation of the reference road surface is calculated for each longitudinal profile. Combined with the maximum depth represented by linear defects, the longitudinal profile elevation difference segmentation threshold is obtained through a weighted average algorithm or mean value processing.

[0142] In step 503, all points in the second preset area are traversed, and points whose elevation difference with the corresponding position of the reference road surface is greater than the longitudinal segmentation threshold are identified as longitudinal expansion disease points. Based on the longitudinal section elevation difference segmentation threshold, this invention discusses whether each point in the potential fourth preset area is an expansion disease point on a longitudinal section-by-section basis. Specifically, points whose elevation difference is greater than the longitudinal segmentation threshold are marked as longitudinal expansion disease points, and the longitudinal expansion disease points are recorded in the expansion binary map.

[0143] In step 504, after denoising all longitudinally expanding disease points, the average and maximum ranges of the longitudinal disease expansion are determined. Based on the average and maximum ranges of the longitudinal disease expansion, the longitudinal expansion range of the disease is determined. This invention utilizes a morphological processing method, employing a dilation-then-erosion approach to extend the disease. Based on the length characteristics of connected regions, shorter noise regions are removed, thereby determining the denoised longitudinally expanding disease points. The average and maximum ranges of the longitudinal disease expansion are statistically analyzed, and the influence range of the longitudinal disease expansion is calculated. This invention can use a weighted algorithm to process the average and maximum ranges of the longitudinal disease expansion, thereby determining the influence range of the longitudinal disease expansion.

[0144] Figure 6 This is a flowchart illustrating the process of extracting three-dimensional information of pavement defects according to the present invention. The extraction of three-dimensional information of pavement defects based on the defect length, defect width, and defect depth includes:

[0145] When the disease type is linear disease, for any disease point in the disease area, the depth of the disease point is determined by the difference between the elevation of the disease point and the corresponding position elevation of the reference road surface. All disease points in the disease area are traversed to obtain the depth of the disease of all disease points in the disease area. The depth of the disease of all disease points is then averaged to obtain the average depth of the linear disease.

[0146] The depth distribution data is sorted to obtain a depth dataset arranged in ascending or descending order. Based on the depth dataset arranged in ascending or descending order, an initial representative maximum depth is determined as the representative maximum depth of the linear disease according to the number of depth data and a preset representative maximum depth quantile; or, the mean of depth data in the depth distribution that is greater than the initial representative maximum depth is taken as the representative maximum depth of the linear disease.

[0147] The third weighting value is determined based on the average depth of linear diseases and the third weighting coefficient;

[0148] The fourth weighting value is determined based on the maximum representative depth of linear diseases and the fourth weighting coefficient.

[0149] The depth of influence of linear diseases is determined based on the third and fourth weighted values;

[0150] Three-dimensional information of linear defects in the three-dimensional pavement is extracted based on the length, width, and depth of the defects.

[0151] When the disease type is a sheet disease, the depth of influence of the sheet disease is obtained based on the maximum representative depth of the sheet disease and the preset minimum depth of the sheet disease.

[0152] Based on the length, width, and depth of the surface defects, three-dimensional information of surface defects in the three-dimensional pavement is extracted.

[0153] In step 601, when the disease type is linear disease, for any disease point in the disease area, the depth of the disease point is determined by the difference between the elevation of the disease point and the corresponding elevation of the reference road surface. All disease points in the disease area are traversed to obtain the depth of all disease points in the disease area. The depth of all disease points is averaged to obtain the average depth of the linear disease. The depth information of the coordinates of all points in the linear disease area is obtained by subtracting the elevation of all points in the linear disease area from the corresponding points of the reference plane in its coordinate system. The depths of all points are sorted to obtain the depth distribution of the linear disease area. The depth information of all points in the linear disease area is averaged to obtain the average depth of the linear disease.

[0154] In step 602, referring to step 501, the representative maximum depth can be calculated according to a preset representative maximum depth quantile. Specifically, the depth distribution data is sorted to obtain a depth dataset arranged in ascending or descending order. Based on the depth dataset arranged in ascending or descending order, the initial representative maximum depth is determined by multiplying the number of depth data points by the preset representative maximum depth quantile, and this initial representative maximum depth is determined as the representative maximum depth of the linear disease. In another optional embodiment, the initial representative maximum depth is compared with all depth data in the depth distribution, and depth data points greater than the initial representative maximum depth are identified. All depth data points greater than the initial representative maximum depth are averaged, and the averaged result is determined as the representative maximum depth of the linear disease.

[0155] In step 603, a third weighting value is determined based on the product of the average depth of the linear disease and the third weighting coefficient, wherein the third weighting coefficient is a preset weighting coefficient.

[0156] In step 604, a fourth weighting value is determined based on the product of the maximum representative depth of the linear disease and the fourth weighting coefficient. The fourth weighting coefficient is a preset weighting coefficient, and the sum of the third weighting coefficient and the fourth weighting coefficient is a preset constant. Optionally, the preset constant can be 1.

[0157] In step 605, the depth of influence of the linear disease is determined based on the sum of the third weighted value and the fourth weighted value. For example, if the linear disease is linear, the depth of influence of the linear disease can be determined by referring to the following formula:

[0158] DR=W D1 *DA+W D2 *DM(2)

[0159] Wherein, DR represents the depth of influence of linear diseases, and W D1 W is the third weighting coefficient. D2 DA is the fourth weighting coefficient, DM is the average depth of linear diseases, and DA is the maximum representative depth of linear diseases.

[0160] In step 606, three-dimensional information of linear defects in the three-dimensional pavement is extracted based on the defect length, defect width, and linear defect depth. For linear defect areas, the lateral and longitudinal expansion ranges of the defects are not set. The defect length represents the maximum length, and the defect width represents the maximum width. In other embodiments, the lateral expansion range of the defects can be set to a first preset constant, and the longitudinal expansion range of the defects can be set to a second preset constant.

[0161] Those skilled in the art will understand that, in the case of a disease type that is a sheet disease, the depth of influence of the sheet disease is obtained based on the maximum representative depth of the sheet disease and the preset minimum depth of the sheet disease.

[0162] Based on the length, width, and depth of the surface defects, three-dimensional information of surface defects in the three-dimensional pavement is extracted.

[0163] Those skilled in the art will understand that, in an optional embodiment, the present invention is based on […] for areas of planar disease. Figure 4 as well as Figure 5The lateral and longitudinal expansion ranges of the disease were determined. If the length direction of the areal disease area is lateral, the disease impact length is the sum of the maximum representative length and the lateral expansion range, and the disease impact width is the sum of the maximum representative width and the longitudinal expansion range. If the length direction of the areal disease area is longitudinal, the disease impact length is the sum of the maximum representative length and the longitudinal expansion range, and the disease impact width is the sum of the maximum representative width and the lateral expansion range. If the areal disease area has a significant depth, the disease impact depth is determined by the maximum representative depth of the areal disease and a preset minimum depth of the areal disease; otherwise, the disease impact depth is the maximum representative depth of the areal disease.

[0164] Figure 7 This is a flowchart illustrating the method for obtaining the types of road defects based on precise three-dimensional pavement. Before determining the corresponding defect influence length and width for each defect type based on its representative length, representative width, lateral expansion range, and longitudinal expansion range, the method further includes:

[0165] The disease ratio is determined based on the length and width of the circumscribed rectangle of the diseased area and the area of ​​the diseased area.

[0166] If the disease ratio is greater than or equal to a preset ratio and the circumscribed width is greater than a preset width, the disease type is determined to be a planar disease.

[0167] If the disease ratio is less than a preset ratio, or the circumscribed width is less than or equal to a preset width, the disease type is determined to be linear disease.

[0168] The surface-like defects include macroscopic deformation defects and / or reticulate defects.

[0169] In step 701, the corresponding circumscribed moment of the diseased area is determined, and the length of the circumscribed moment is obtained. The area occupied by the diseased area is further determined based on the distribution of all points in the diseased area. The disease ratio is determined based on the quotient of the length of the circumscribed moment of the diseased area and the area of ​​the diseased area.

[0170] In step 702, the preset ratio can be 0.05, 0.1, 0.15, etc., and the preset width can be 0.1, 0.2, 0.3 (unit: meters), etc. If the disease ratio is determined to be 0.6 and the width of the circumscribed moment of the diseased area is 0.4 based on the length and width of the circumscribed moment of the diseased area and the area of ​​the diseased area, then the disease type of the diseased area is determined to be a planar disease.

[0171] In step 703, if the disease ratio is determined to be 0.12 and the width of the circumscribed moment of the diseased area is 0.3 based on the length and width of the circumscribed moment of the diseased area and the area of ​​the diseased area, and the preset ratio is 0.15, then the disease type of the diseased area is determined to be linear disease.

[0172] The surface-like defects include macroscopic deformation defects and / or reticulate defects.

[0173] This invention also provides a computer program product, comprising a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a three-dimensional pavement distress extraction method provided by the methods described above. This method includes: obtaining representative length, representative width, lateral extension range, and longitudinal extension range for each distress type based on precise three-dimensional pavement data and distress type information; determining the corresponding distress influence length and width for each distress type based on the representative length, representative width, lateral extension range, and longitudinal extension range; extracting three-dimensional information of the pavement distress based on the distress influence length, distress influence width, and distress influence depth; and, in the case of linear distress, [further details on the method would be needed]. The representative length is determined based on the diagonal length of the circumscribed moment of the linear disease, and the representative width is determined based on the average width of the linear disease and the maximum representative width of the linear disease. The lateral expansion range of the disease is a first preset constant, and the longitudinal expansion range of the disease is a second preset constant. In the case of a planar disease, the representative length is determined based on the maximum representative length of the planar disease, and the representative width is determined based on the maximum representative width of the planar disease. The lateral expansion range of the disease is the lateral expansion disease points selected from the first preset region based on a lateral segmentation threshold, and the longitudinal expansion range of the disease is the longitudinal expansion disease points selected from the first preset region based on a longitudinal segmentation threshold. The first preset region is the area centered on the disease area location point, within a first preset radius, that does not overlap with the disease area.

[0174] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the three-dimensional pavement distress extraction method provided by the above methods. This method includes: obtaining, based on precise three-dimensional pavement data and distress type information, a representative length, representative width, lateral extension range, and longitudinal extension range of each distress type; determining the distress impact length and distress impact width corresponding to each distress type based on the representative length, representative width, lateral extension range, and longitudinal extension range of each distress type; extracting three-dimensional information of the pavement distress based on the distress impact length, distress impact width, and distress impact depth; and, in the case of linear distress, the representative length is based on the linear distress... The length of the circumscribed diagonal is determined, and the representative width is determined based on the average width of linear diseases and the maximum representative width of linear diseases. The lateral expansion range of the disease is a first preset constant, and the longitudinal expansion range of the disease is a second preset constant. In the case of a planar disease, the representative length is determined based on the maximum representative length of the planar disease, and the representative width is determined based on the maximum representative width of the planar disease. The lateral expansion range of the disease is the lateral expansion disease points selected from the first preset region based on the lateral segmentation threshold, and the longitudinal expansion range of the disease is the longitudinal expansion disease points selected from the first preset region based on the longitudinal segmentation threshold. The first preset region is the region that does not overlap with the disease region within a first preset radius, centered on the disease region's location point.

[0175] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0176] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0177] 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 extracting three-dimensional information on road surface defects based on precise three-dimensional pavement, characterized in that, include: Based on precise three-dimensional pavement data and disease type information, obtain the representative length, representative width, lateral expansion range, and longitudinal expansion range of each disease type. The length and width of each disease type are determined based on its representative length, representative width, horizontal expansion range, and vertical expansion range. Three-dimensional information of pavement defects is extracted based on the length, width, and depth of the defects. When the disease type is linear, the representative length is determined based on the length of the diagonal of the circumscribed rectangle of the linear disease, and the representative width is determined based on the average width of the linear disease and the maximum representative width of the linear disease, including: For any disease point, take the disease point as the center and obtain the representative direction of the set of all points in the third preset area; If the angle between the representative direction and the lateral direction is less than or equal to a preset angle, count the number of consecutive points in the longitudinal direction at the location of the disease point, and obtain the width of the disease point based on the number of consecutive points in the longitudinal direction and the longitudinal sampling interval; otherwise, count the number of consecutive points in the lateral direction at the location of the disease point, and obtain the width of the disease point based on the number of consecutive points in the lateral direction and the lateral sampling interval. Traverse all disease points and obtain the width of each disease point to determine the average width of the linear disease based on the width of each disease point. The number of width data points is determined based on the number of disease points in the diseased area. The initial representative maximum width, determined based on the number of width data points and the preset representative maximum width quantile, is taken as the representative maximum width of the linear disease; or, the average width data in the width distribution that is greater than the initial representative maximum width is taken as the representative maximum width of the linear disease. The first weighting value is determined based on the average width of the linear disease and the first weighting coefficient. The second weighting value is determined based on the maximum width represented by linear diseases and the second weighting coefficient; The representative width is determined based on the first weighted value and the second weighted value: WR=W W1 *WA+W W2 *WM Where WR is the representative width, W W1 W is the first weighting coefficient. W2 The second weighting coefficient is WA, which is the average width of the linear disease, and WM is the maximum representative width of the linear disease. The sum of the first weighting coefficient and the second weighting coefficient is a third preset constant; the lateral expansion range of the disease is a first preset constant, and the longitudinal expansion range of the disease is a second preset constant; When the disease type is a planar disease, the representative length is determined based on the maximum representative length of the planar disease, the representative width is determined based on the maximum representative width of the planar disease, the lateral expansion range of the disease is the lateral expansion disease points selected from the first preset region based on the lateral segmentation threshold, and the longitudinal expansion range of the disease is the longitudinal expansion disease points selected from the first preset region based on the longitudinal segmentation threshold. The first preset area is the area that does not overlap with the diseased area within a first preset radius, centered on the location point of the diseased area.

2. The method for extracting three-dimensional information on road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, Before determining the length and width of the disease impact corresponding to each disease type, the following steps are also included: After extracting the second preset area and the defect area from the total road surface area, the ideal road surface area is obtained, and the reference road surface is filtered out based on the ideal road surface area. The second preset area is the area that does not overlap with the diseased area within a second preset radius, centered on the location point of the diseased area.

3. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, The representative length is determined based on the length of the diagonal of the circumscribed moment of the linear disease, including: Obtain the length and width of the circumscribed radius of the linear disease area; The diagonal length of the circumscribed moment is calculated based on its length and width to obtain the representative length of the linear disease.

4. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, The representative length is determined based on the maximum representative length of the areal disease, including: The number of length data points is determined based on the location information of the area of ​​the disease, and the initial representative maximum length, determined based on the number of length data points and the preset representative maximum length quantile, is taken as the representative maximum length of the disease. Alternatively, the average length of the length data in the length distribution that is greater than the initial representative maximum length can be used as the representative maximum length of the areal disease.

5. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, The representative width is determined based on the maximum representative width of the area disease, including: The number of width data points is determined based on the location information of the area of ​​the disease, and the initial representative maximum width, determined based on the number of width data points and the preset representative maximum width quantile, is used as the representative maximum width of the disease. Alternatively, the average width of the width data in the width distribution that is greater than the initial representative maximum width can be used as the representative maximum width of the areal disease.

6. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 2, characterized in that, The lateral expansion range of the disease is determined from the laterally expanding disease points selected from the first preset region based on a lateral segmentation threshold, including: The depth distribution and number of depth data points are determined based on the depth of all disease points within the area of ​​the disease. The initial representative maximum depth, determined based on the number of depth data points and a preset representative maximum depth quantile, is taken as the representative maximum depth of the disease. Alternatively, the average depth data points in the depth distribution that are greater than the initial representative maximum depth are taken as the representative maximum depth of the disease. The fourth preset area where the disease area is located is determined. The elevation difference between any point in the fourth preset area and the corresponding position of the reference road surface is obtained by traversing all points in the fourth preset area. The lateral segmentation threshold is determined based on the average of all elevation differences and the maximum representative depth of the surface disease. Traverse all points in the second preset area, and determine the points whose elevation difference between any point and the corresponding position of the reference road surface is greater than the horizontal segmentation threshold as horizontal expansion disease points; After denoising all laterally expanding disease points, determine the average and maximum range of lateral disease expansion, and then determine the lateral expansion range of the disease based on the average and maximum range of lateral disease expansion. The fourth preset region is the region in the cross-sectional data where the diseased region is located, after removing the diseased region and the second preset region.

7. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 2, characterized in that, The longitudinal expansion range of the disease is determined from the longitudinally expanding disease points selected from the first preset region based on the longitudinal segmentation threshold, including: The depth distribution and number of depth data points are determined based on the depth of all disease points within the area of ​​the disease. The initial representative maximum depth, determined based on the number of depth data points and a preset representative maximum depth quantile, is taken as the representative maximum depth of the disease. Alternatively, the average depth data points in the depth distribution that are greater than the initial representative maximum depth are taken as the representative maximum depth of the disease. The fifth preset area where the disease area is located is determined. For any point in the fifth preset area, the elevation difference between the elevation of the point and the corresponding position of the reference road surface is obtained by traversing all points in the fifth preset area. The longitudinal segmentation threshold is determined based on the average of all elevation differences and the maximum representative depth of the surface disease. Traverse all points in the second preset area, and determine the points whose elevation difference between any point and the corresponding position of the reference road surface is greater than the longitudinal segmentation threshold as longitudinal expansion disease points; After denoising all longitudinally expanding disease points, determine the average and maximum range of longitudinal disease expansion, and then determine the longitudinal expansion range of the disease based on the average and maximum range of the longitudinal disease expansion. The fifth preset region is the region in the longitudinal section data where the diseased region is located, after removing the diseased region and the second preset region.

8. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, The extraction of three-dimensional information of pavement defects based on the length, width, and depth of the defect impact includes: When the disease type is linear disease, for any disease point in the disease area, the depth of the disease point is determined by the difference between the elevation of the disease point and the corresponding position elevation of the reference road surface. All disease points in the disease area are traversed to obtain the depth of the disease of all disease points in the disease area. The depth of the disease of all disease points is then averaged to obtain the average depth of the linear disease. The depth distribution and number of depth data points are determined based on the depth of all disease points within the disease area of ​​the linear disease. The initial representative maximum depth is determined based on the number of depth data points and a preset representative maximum depth quantile as the representative maximum depth of the linear disease; or, the average of the depth data points in the depth distribution that are greater than the initial representative maximum depth is taken as the representative maximum depth of the linear disease. The third weighting value is determined based on the average depth of linear diseases and the third weighting coefficient; The fourth weighting value is determined based on the maximum representative depth of linear diseases and the fourth weighting coefficient. The depth of influence of linear diseases is determined based on the third and fourth weighted values; Three-dimensional information of linear defects in the three-dimensional pavement is extracted based on the length, width, and depth of the defects. When the disease type is a sheet disease, the depth of influence of the sheet disease is obtained based on the maximum representative depth of the sheet disease and the preset minimum depth of the sheet disease. Based on the length, width, and depth of the surface defects, three-dimensional information of surface defects in the three-dimensional pavement is extracted.

9. The method for extracting three-dimensional information of road defects based on precise three-dimensional pavement as described in claim 1, characterized in that, Before determining the corresponding disease impact length and width for each disease type based on its representative length, representative width, lateral expansion range, and longitudinal expansion range, the process also includes: The disease ratio is determined based on the length and width of the circumscribed rectangle of the diseased area and the area of ​​the diseased area. If the disease ratio is greater than or equal to a preset ratio and the circumscribed width is greater than a preset width, the disease type is determined to be a planar disease. If the disease ratio is less than a preset ratio, or the circumscribed width is less than or equal to a preset width, the disease type is determined to be linear disease. The surface-like defects include macroscopic deformation defects and / or reticulate defects.

Images