Roadway centerline acquisition method

Abstract

The invention provides a roadway center line obtaining method, and relates to the technical field of roadway tunneling, and the method comprises the steps: obtaining a roadway point cloud model of a to-be-calibrated roadway; generating a plurality of roadway point cloud slices distributed along the principal component direction of the roadway point cloud model based on the point cloud slice spacing and the point cloud slice thickness; dividing each roadway point cloud slice into a plurality of grids, taking the corresponding grid when the point cloud density in each roadway point cloud slice is greater than a preset density threshold as a high-density grid, and determining a center point coordinate corresponding to each roadway point cloud slice based on a centroid coordinate of each high-density grid; and performing curve fitting on the center point corresponding to each roadway point cloud slice based on each center point coordinate to obtain a roadway center line of the to-be-calibrated roadway. According to the invention, the central point of each roadway point cloud slice is subjected to curve fitting, so that the roadway center line is obtained, and the efficiency and accuracy of obtaining the roadway center line are improved.

Description

Technical Field

[0001] This invention relates to the field of tunnel excavation technology, and in particular to a method for obtaining the centerline of a tunnel. Background Technology

[0002] In the process of mine mining, roadway excavation is one of the key links to ensure safe and efficient mine production. In order to ensure the accuracy and efficiency of roadway excavation, it is necessary to accurately determine the centerline of the roadway. The traditional method of roadway centerline marking mainly involves manual on-site measurement and marking using tools such as slope gauges and total stations. The existing centerline marking method requires setting up multiple total stations in the roadway. Before centerline marking, it is necessary to manually check whether the total station is stable. In the process of marking complex roadways, it is necessary to manually calculate the segment chord length and turning angle. After marking the centerline, it is necessary to manually suspend a plumb bob and check the collinearity of the three points. Therefore, the existing manual centerline marking process has technical problems of low efficiency and low accuracy. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method for obtaining the centerline of a roadway, which can solve the technical problems of low efficiency and accuracy in the prior art when the roadway is calibrated manually.

[0004] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows: In a first aspect, embodiments of the present invention provide a method for obtaining the centerline of a roadway, comprising: Obtain the point cloud model of the tunnel to be calibrated; wherein the point cloud model of the tunnel is located in the point cloud coordinate system; Multiple lane point cloud slices distributed along the principal component direction of the lane point cloud model are generated based on the point cloud slice spacing and point cloud slice thickness. Each of the roadway point cloud slices is divided into multiple grids. The grids in each roadway point cloud slice where the point cloud density is greater than a preset density threshold are taken as high-density grids. The center point coordinates of each roadway point cloud slice are determined based on the centroid coordinates of each high-density grid. Based on the coordinates of each center point, curve fitting is performed on the center points corresponding to each tunnel point cloud slice to obtain the tunnel centerline of the tunnel to be calibrated.

[0005] Furthermore, this embodiment of the invention provides a first possible implementation of the first aspect, wherein the step of generating multiple tunnel point cloud slices distributed along the principal component direction of the tunnel point cloud model based on the point cloud slice spacing and point cloud slice thickness includes: The point clouds in the tunnel point cloud model are preprocessed to obtain the preprocessed tunnel point cloud model; wherein, the preprocessing includes: denoising processing and downsampling processing; Principal component analysis is performed on the preprocessed tunnel point cloud model to determine the principal component orientation of the tunnel point cloud model; Based on the point cloud slice spacing and the point cloud slice thickness, multiple lane point cloud slices are generated that are distributed along the principal component direction of the lane point cloud model.

[0006] Furthermore, this embodiment of the invention provides a second possible implementation of the first aspect, wherein the step of preprocessing each of the point clouds in the lane point cloud model includes: The roadway point cloud model is denoised using the statistical outlier method to remove outlier point clouds from the model, resulting in a denoised roadway point cloud model. The denoised tunnel point cloud model is downsampled using a voxel grid filter with a preset voxel size to obtain the preprocessed tunnel point cloud model.

[0007] Furthermore, this embodiment of the invention provides a third possible implementation of the first aspect, wherein the step of performing principal component analysis on the preprocessed tunnel point cloud model to determine the principal component orientation of the tunnel point cloud model includes: Obtain the centroid coordinates of the preprocessed tunnel point cloud model, center the coordinates of each point cloud in the tunnel point cloud model based on the centroid coordinates, and construct the point cloud matrix corresponding to the tunnel point cloud model based on the coordinates of each point cloud after centering. Based on the point cloud matrix, the covariance matrix corresponding to the tunnel point cloud model is determined, and multiple eigenvalues ​​are obtained by performing feature analysis on the covariance matrix. The direction represented by the eigenvector corresponding to the largest eigenvalue is taken as the principal component direction of the tunnel point cloud model.

[0008] Furthermore, this embodiment of the invention provides a fourth possible implementation of the first aspect, wherein the step of dividing each of the roadway point cloud slices into multiple grids, designating the grids in each roadway point cloud slice where the point cloud density is greater than a preset density threshold as high-density grids, and determining the center point coordinates of each roadway point cloud slice based on the centroid coordinates of each high-density grid includes: Establish a slice coordinate system corresponding to each of the roadway point cloud slices, and transform the coordinates of the point cloud contained in each of the point cloud slices from the point cloud coordinate system to the slice coordinate system; Each of the aforementioned tunnel point cloud slices is divided into multiple grids, and the point cloud density corresponding to each grid is determined based on the number of point clouds contained in each grid. Determine the high-density grid corresponding to the point cloud density that is greater than a preset density threshold in each of the roadway point cloud slices; The center point coordinates of each roadway point cloud slice are determined based on the point cloud density corresponding to each high-density grid in each roadway point cloud slice and the centroid coordinates of each high-density grid; wherein the centroid coordinates of the high-density grid and the center point coordinates are located in the slice coordinate system.

[0009] Furthermore, this embodiment of the invention provides a fifth possible implementation of the first aspect, wherein the step of establishing the slice coordinate system corresponding to each of the tunnel point cloud slices includes: The orientation unit normal vector of the roadway point cloud model is determined based on the three-dimensional coordinates of the preset starting point cloud and the preset ending point cloud in the point cloud coordinate system; wherein, the direction of the orientation unit normal vector is the same as the direction of the principal component. Obtain the corresponding gravity unit normal vector in the world coordinate system, and determine the vertical axis unit normal vector in the slice coordinate system corresponding to each of the tunnel point cloud slices based on the orientation unit normal vector and the gravity unit normal vector; Based on the directional unit normal vector and the vertical axis unit normal vector, determine the horizontal axis unit normal vector in the slice coordinate system corresponding to each of the roadway point cloud slices; Based on the preset origin point cloud in each of the tunnel point cloud slices, the direction of the horizontal axis unit normal vector, and the direction of the vertical axis unit normal vector, a slice coordinate system corresponding to each of the tunnel point cloud slices is constructed.

[0010] Furthermore, this embodiment of the invention provides a sixth possible implementation of the first aspect, wherein the step of transforming the coordinates of the point cloud contained in each point cloud slice from the point cloud coordinate system to the slice coordinate system includes: The vector between each point cloud and the preset origin point cloud is determined based on the preset origin point cloud corresponding to each of the tunnel point cloud slices and the three-dimensional coordinates of each point cloud in the point cloud coordinate system; The two-dimensional coordinates of each point cloud in the corresponding slice coordinate system are determined based on the vector between each point cloud and the preset origin point cloud, the horizontal axis unit normal vector, and the vertical axis unit normal vector.

[0011] Furthermore, this embodiment of the invention provides a seventh possible implementation of the first aspect, wherein the step of determining the center point coordinates corresponding to each of the roadway point cloud slices based on the point cloud density corresponding to each of the high-density grids in each of the roadway point cloud slices and the centroid coordinates of each of the high-density grids includes: Each high-density grid in each row of the grid of each tunnel point cloud slice is divided into a left high-density grid and a right high-density grid. The centroid coordinates of the left side of each row of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding left high-density grid in each row of grids; the centroid coordinates of the right side of each row of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding right high-density grid in each row of grids. The center point coordinates of each tunnel point cloud slice are determined based on the left and right centroid coordinates of each row of grids.

[0012] Furthermore, this embodiment of the invention provides an eighth possible implementation of the first aspect, wherein the step of determining the center point coordinates corresponding to each of the roadway point cloud slices based on the point cloud density corresponding to each of the high-density grids in each of the roadway point cloud slices and the centroid coordinates of each of the high-density grids includes: Each high-density grid in each column of the grid of each of the aforementioned roadway point cloud slices is divided into an upper high-density grid and a lower high-density grid; The centroid coordinates of the upper side of each column of grid are determined based on the point cloud density, center x-coordinate, and center y-coordinate of the corresponding upper high-density grid in each column of grid; the centroid coordinates of the lower side of each column of grid are determined based on the point cloud density, center x-coordinate, and center y-coordinate of the corresponding lower high-density grid in each row of grid. The center point coordinates of each tunnel point cloud slice are determined based on the upper and lower centroid coordinates of each column of the grid.

[0013] Furthermore, this embodiment of the invention provides a ninth possible implementation of the first aspect, wherein the step of performing curve fitting on the center points corresponding to each of the roadway point cloud slices based on the coordinates of each of the center points to obtain the roadway centerline of the roadway to be calibrated includes: The coordinates of the center point corresponding to each of the roadway point cloud slices are transformed from the slice coordinate system to the point cloud coordinate system; Based on the curve fitting method and the coordinates of each center point in the point cloud coordinate system, the center point corresponding to each roadway point cloud slice is curve fitted to obtain the roadway centerline to be calibrated; wherein, the curve fitting method includes the B-spline curve fitting method.

[0014] This invention provides a method for obtaining the centerline of a roadway. The method includes: obtaining a point cloud model of the roadway to be calibrated; wherein the point cloud model is located in a point cloud coordinate system; generating multiple roadway point cloud slices distributed along the principal component direction of the roadway point cloud model based on the point cloud slice spacing and point cloud slice thickness; dividing each roadway point cloud slice into multiple grids, and taking the grids corresponding to the point cloud density of each roadway point cloud slice that are greater than a preset density threshold as high-density grids, determining the center point coordinates of each roadway point cloud slice based on the centroid coordinates of each high-density grid; and performing curve fitting on the center points corresponding to each roadway point cloud slice based on the center point coordinates to obtain the roadway centerline of the roadway to be calibrated. This invention obtains a point cloud model of the roadway to be calibrated, slices the model based on the point cloud slice spacing and thickness, generating multiple point cloud slices distributed along the principal component direction. Then, each slice is divided into multiple grids, and a high-density grid is determined based on the point cloud density. Higher point cloud density indicates a closer proximity to the roadway's sidewall region. The centroid coordinates of the high-density grid accurately determine the coordinates of the center point corresponding to each slice. Finally, the center point of each slice is determined using these coordinates, and curve fitting is performed on each center point to obtain the roadway centerline. This avoids the inefficiency and low accuracy of manual centerline calibration in existing technologies. This method improves the accuracy of determining center point coordinates by slicing the roadway point cloud model, ultimately determining each center point based on its coordinates, and finally, curve fitting on each center point allows for rapid and accurate determination of the roadway centerline.

[0015] Other features and advantages of the embodiments of the present invention will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above in the embodiments of the present invention.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 The diagram illustrates the steps of a method for obtaining the centerline of a roadway provided by an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be described below in conjunction with the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0020] This embodiment provides a method for obtaining the centerline of a roadway. For details, please refer to... Figure 1 The diagram shows the steps of a method for obtaining the centerline of a roadway. This method mainly includes: S10, Obtain the point cloud model of the roadway to be calibrated; wherein, the point cloud model of the roadway is located in the point cloud coordinate system; Point cloud data in the roadway to be calibrated is collected using point cloud acquisition equipment (including an explosion-proof 3D laser scanner). Principal component analysis is then performed on the roadway point cloud model to obtain the first eigenvalue. Second eigenvalue and the third eigenvalue ,and The larger the eigenvalue, the more dispersed the point cloud data. Therefore, the direction represented by the first eigenvector corresponding to the first eigenvalue (i.e., the principal component direction) is considered to be the direction of roadway extension (i.e., the roadway direction), the direction represented by the second eigenvector corresponding to the second eigenvalue is considered to be the direction of roadway width, and the direction represented by the third eigenvector corresponding to the third eigenvalue is considered to be the direction of roadway height. The direction represented by the first eigenvector is taken as the x-axis direction in the point cloud coordinate system, the direction represented by the second eigenvector is taken as the y-axis direction in the point cloud coordinate system, and the direction represented by the third eigenvector is taken as the z-axis direction in the point cloud coordinate system. A point cloud coordinate system is constructed based on the origin, the x-axis direction, the y-axis direction, and the z-axis direction of the point cloud coordinate system. A roadway point cloud model located in the point cloud coordinate system is generated based on the point cloud data.

[0021] S20, Based on the point cloud slice spacing and point cloud slice thickness, generate multiple roadway point cloud slices distributed along the principal component direction of the roadway point cloud model; Multiple lane point cloud slices are generated based on the point cloud slice spacing and point cloud slice thickness, distributed along the principal component direction of the lane point cloud model. The point cloud slice spacing represents the distance between two adjacent lane point cloud slices (i.e., how many distances are between each lane point cloud slice). The size of the point cloud slice spacing is related to the subsequent calculation accuracy. To maintain high calculation accuracy, the point cloud slice spacing should be set smaller. The point cloud slice thickness represents the slice thickness of each lane point cloud slice. Specifically, multiple lane point cloud slices are generated based on the point cloud slice spacing (usually set to 0.5m-2m) and the point cloud slice thickness (usually set to 0.05m-0.1m), distributed along the principal component direction of the lane point cloud model (i.e., the lane extension direction, the direction represented by the first eigenvector mentioned above).

[0022] S30, divide each roadway point cloud slice into multiple grids, and take the grid corresponding to the point cloud density in each roadway point cloud slice when the point cloud density is greater than the preset density threshold as a high-density grid, and determine the center point coordinates of each roadway point cloud slice based on the centroid coordinates of each high-density grid. Each tunnel point cloud slice is divided into multiple grids, and the grids in each tunnel point cloud slice are of the same size. Each grid in each tunnel point cloud slice contains a different number of points. The more points in each grid, the higher the point cloud density of that grid. When the point cloud density is greater than a preset density threshold, it is considered a high-density grid. Since the dense area of ​​point cloud in the tunnel point cloud model usually represents the side wall area of ​​the tunnel, the high-density grid usually represents the side wall of the tunnel. The centroid coordinates of the high-density grid are determined, and the center point coordinates of each tunnel point cloud slice are determined.

[0023] S40, based on the coordinates of each center point, curve fitting is performed on the center points corresponding to each roadway point cloud slice to obtain the roadway centerline of the roadway to be calibrated; The center point of each roadway point cloud slice is determined based on the coordinates of the center point of each roadway point cloud slice. Since the roadway point cloud slices are arranged sequentially along the principal component direction, the center point of each roadway point cloud slice is sequentially fitted based on the curve fitting algorithm to eliminate the sawtooth fluctuation of the roadway centerline caused by point cloud noise and roadway point cloud slice calculation error, and a smooth roadway centerline is obtained.

[0024] The method for obtaining the centerline of a roadway provided by this invention acquires a point cloud model of the roadway to be calibrated, slices the point cloud model based on the point cloud slice spacing and thickness, and generates multiple point cloud slices distributed along the principal component direction. Then, each point cloud slice is divided into multiple grids, and a high-density grid is determined based on the point cloud density. The higher the point cloud density, the closer it is to the sidewall region of the roadway. The centroid coordinates of the high-density grid can accurately determine the coordinates of the center point corresponding to each point cloud slice. Finally, the center point corresponding to each point cloud slice is determined by the center point coordinates, and curve fitting is performed on each center point to obtain the roadway centerline. This method avoids the technical problems of low efficiency and low accuracy when manually calibrating the centerline in the prior art. This method improves the accuracy of determining the center point coordinates by slicing the point cloud model. Finally, each center point is determined based on the coordinates of each center point, and curve fitting is performed on each center point to quickly and accurately determine the roadway centerline.

[0025] In one embodiment, this embodiment provides a specific implementation method for generating multiple tunnel point cloud slices distributed along the principal component direction of the tunnel point cloud model based on the point cloud slice spacing and point cloud slice thickness: S201, preprocess each point cloud in the tunnel point cloud model to obtain the preprocessed tunnel point cloud model; wherein, the preprocessing includes: noise reduction processing and downsampling processing; S203, perform principal component analysis on the preprocessed tunnel point cloud model to determine the principal component direction of the tunnel point cloud model; S205, Based on the point cloud slice spacing and point cloud slice thickness, generate multiple roadway point cloud slices distributed along the principal component direction of the roadway point cloud model; The statistical outlier method is used to denoise the point clouds in the roadway point cloud model, removing point clouds that do not belong to the roadway structure (such as personnel, equipment, vehicles, and fly noise points), thereby reducing the point cloud noise in the roadway point cloud model. The voxel grid filter is used to downsample the roadway point cloud model, reducing the number of point clouds and uniformly reducing the point cloud density while maintaining the geometric features of the roadway point cloud model, thus improving the processing speed of subsequent processing of the roadway point cloud model. Since the point cloud data contained in the tunnel point cloud model is usually large in quantity and randomly distributed, it is impossible to determine the tunnel direction. Therefore, principal component analysis is performed on the preprocessed tunnel point cloud model. The direction with the greatest dispersion of the point cloud data is used to determine the principal component direction of the tunnel point cloud model. This can accurately determine the approximate extension direction of the tunnel point cloud model. Finally, based on the point cloud slice spacing and point cloud slice thickness, multiple tunnel point cloud slices distributed along the principal component direction (i.e. extension direction) of the tunnel point cloud model are generated.

[0026] In one embodiment, this embodiment provides a specific implementation method for preprocessing each point cloud in the lane point cloud model: S2011, The roadway point cloud model is denoised based on the statistical outlier method. Outlier point clouds in the roadway point cloud model are removed to obtain the denoised roadway point cloud model. The denoising process for the roadway point cloud model is based on the statistical outlier method. Specifically, the average distance between each point cloud and its k nearest neighbors is calculated. The mean and standard deviation of the average distances for all point clouds in the roadway point cloud model are calculated. An average distance threshold is set based on the mean and standard deviation of the average distances (usually set to mean ± 2 × standard deviation of average distance). If the average distance of a point cloud is greater than or less than the average distance threshold, the point cloud is identified as an outlier and removed, thus completing the denoising process for the roadway point cloud model.

[0027] S2013, The denoised tunnel point cloud model is downsampled based on a voxel grid filter with a preset voxel size to obtain a preprocessed tunnel point cloud model. The denoised tunnel point cloud model is downsampled using a voxel grid filter with a preset voxel size. Specifically, the voxel grid filter with the preset voxel size represents all points falling within the same voxel in the tunnel point cloud model using the voxel center point, reducing the amount of point cloud data. The preset voxel size is... Preferably, it is a cube with length, width and height of 1mm.

[0028] In one embodiment, this embodiment provides a specific implementation method for performing principal component analysis on a preprocessed tunnel point cloud model to determine the principal component orientation of the tunnel point cloud model: S2031, obtain the centroid coordinates of the preprocessed tunnel point cloud model, center the coordinates of each point cloud in the tunnel point cloud model based on the centroid coordinates, and construct the point cloud matrix corresponding to the tunnel point cloud model based on the coordinates of each point cloud after centering. The average value of each coordinate component is determined based on the three-dimensional coordinates of all points in the preprocessed tunnel point cloud model. Specifically, the average value of the horizontal coordinates of all points is obtained by summing the values ​​of the horizontal coordinates of all points and dividing by the number of points in the preprocessed tunnel point cloud model. The average value of the vertical coordinates of all points is obtained by summing the values ​​of the vertical coordinates of all points and dividing by the number of points in the preprocessed tunnel point cloud model. The average value of the vertical coordinates of all points is obtained by summing the values ​​of the vertical coordinates of all points and dividing by the number of points in the preprocessed tunnel point cloud model. The centroid coordinates are then constructed based on the average values ​​of the horizontal, vertical, and vertical coordinates. Based on the centroid coordinates, the three-dimensional coordinates of all points in the preprocessed tunnel point cloud model are centered. Specifically, the three-dimensional coordinates of all points are subtracted from the centroid coordinates to obtain the centered three-dimensional coordinates of all points. The point cloud matrix corresponding to the tunnel point cloud model is constructed based on the centered three-dimensional coordinates of all points.

[0029] S2033, based on the point cloud matrix, determine the covariance matrix corresponding to the roadway point cloud model, and perform feature analysis on the covariance matrix to obtain multiple eigenvalues. The direction represented by the eigenvector corresponding to the largest eigenvalue is taken as the principal component direction of the roadway point cloud model. Obtain the transpose of the point cloud matrix. Based on the point cloud matrix and its transpose, determine the covariance matrix corresponding to the tunnel point cloud model. Perform eigenvalue decomposition on the covariance matrix to obtain the first eigenvalue. Second eigenvalue and the third eigenvalue ,and Determine the first eigenvalue respectively Second eigenvalue Third eigenvalue The corresponding first feature vector, second feature vector, and third feature vector are used as the principal component directions of the alleyway point cloud model, with the direction represented by the first feature vector being used as the principal component direction.

[0030] In one embodiment, this embodiment provides a specific implementation method for dividing each roadway point cloud slice into multiple grids, designating the grids in each roadway point cloud slice where the point cloud density is greater than a preset density threshold as high-density grids, and determining the center point coordinates of each roadway point cloud slice based on the centroid coordinates of each high-density grid: S301, Establish the slice coordinate system corresponding to each roadway point cloud slice, and transform the coordinates of the point cloud contained in each point cloud slice from the point cloud coordinate system to the slice coordinate system. S303, divide the point cloud slices of each roadway into multiple grids, and determine the point cloud density corresponding to each grid based on the number of point clouds contained in each grid; S305, determine the high-density grid corresponding to the point cloud density in each roadway point cloud slice when the point cloud density is greater than the preset density threshold; S307, Based on the point cloud density corresponding to each high-density grid in each roadway point cloud slice and the centroid coordinates of each high-density grid, determine the center point coordinates corresponding to each roadway point cloud slice; wherein, the centroid coordinates and center point coordinates of the high-density grid are located in the slice coordinate system; Establish a slice coordinate system corresponding to each roadway point cloud slice, and project the point cloud contained in each point cloud slice from the three-dimensional coordinate system (i.e., the point cloud coordinate system) to the two-dimensional coordinate system (i.e., the slice coordinate system) to obtain the two-dimensional point set corresponding to each roadway point cloud slice. Each tunnel point cloud slice is divided into multiple grids of the same size. Specifically, each tunnel point cloud slice is divided into... 1 grid, of which Let i be the number of grid rows corresponding to the i-th lane point cloud slice. Let i be the number of grid columns corresponding to the i-th lane point cloud slice, i = 1, 2, 3, ..., P. There are a total of P lane point cloud slices. Each grid contains a different number of point clouds. The point cloud density corresponding to each grid is determined based on the number of point clouds contained in each grid. The high-density grids corresponding to the point cloud density in each roadway point cloud slice are determined when the point cloud density in each slice is greater than a preset density threshold (the preset density threshold is usually set to 1.5 times the average density of each roadway point cloud slice). Finally, the center point coordinates of each roadway point cloud slice are determined based on the point cloud density corresponding to each high-density grid in each roadway point cloud slice and the centroid coordinates of each high-density grid.

[0031] In one embodiment, this embodiment provides a specific implementation method for establishing the slice coordinate system corresponding to each tunnel point cloud slice: S3011, Determine the orientation unit normal vector of the tunnel point cloud model based on the three-dimensional coordinates of the preset starting point cloud and the preset ending point cloud in the point cloud coordinate system; wherein, the direction of the orientation unit normal vector is the same as the direction of the principal component. The orientation unit normal vector of the roadway point cloud model is determined based on the three-dimensional coordinates of the preset starting point cloud and the preset ending point cloud in the point cloud coordinate system. Specifically, the line connecting the preset starting point cloud and the preset ending point cloud is parallel to the orientation of the principal component direction of the roadway. The orientation vector is obtained by subtracting the preset starting point cloud and the preset ending point cloud. The ratio of the orientation vector to the magnitude of the orientation vector is calculated to obtain the orientation unit normal vector.

[0032] S3013, obtain the corresponding gravity unit normal vector in the world coordinate system, and determine the vertical axis unit normal vector in the slice coordinate system corresponding to each tunnel point cloud slice based on the directional unit normal vector and the gravity unit normal vector. Based on the gravity unit normal vector in the world coordinate system, the direction of the vertical axis unit normal vector in the slice coordinate system corresponding to the tunnel point cloud slice is obtained. Specifically, the dot product of the gravity unit normal vector and the orientation unit normal vector is calculated, and the dot product is cross-multiplied with the orientation unit normal vector to obtain the projection of the gravity unit normal vector onto the orientation unit normal vector. Finally, the difference between the gravity unit normal vector and the cross product result is calculated and standardized to obtain the vertical axis unit normal vector in the slice coordinate system corresponding to each tunnel point cloud slice.

[0033] S3015, Based on the directional unit normal vector and the vertical axis unit normal vector, determine the horizontal axis unit normal vector in the slice coordinate system corresponding to each roadway point cloud slice; The cross product of the directional unit normal vector and the vertical unit normal vector is used to obtain the horizontal unit normal vector of the slice coordinate system corresponding to each tunnel point cloud slice.

[0034] S3017, construct the slice coordinate system corresponding to each roadway point cloud slice based on the preset origin point cloud, the direction of the horizontal axis unit normal vector, and the direction of the vertical axis unit normal vector in each roadway point cloud slice. The preset origin point cloud in each roadway point cloud slice is used as the origin of the slice coordinate system. The direction of the horizontal axis unit normal vector is used as the horizontal axis direction of the slice coordinate system, and the direction of the vertical axis unit normal vector is used as the vertical axis direction of the slice coordinate system. Based on the origin, horizontal axis direction, and vertical axis direction of the slice coordinate system, the slice coordinate system corresponding to each roadway point cloud slice is constructed.

[0035] In one embodiment, this embodiment provides a specific implementation method for transforming the coordinates of the point clouds contained in each point cloud slice from the point cloud coordinate system to the slice coordinate system: S3012, Based on the preset origin point cloud corresponding to each tunnel point cloud slice and the three-dimensional coordinates of each point cloud in the point cloud coordinate system, determine the vector between each point cloud and the preset origin point cloud; Obtain the three-dimensional coordinates of the preset origin point cloud corresponding to each tunnel point cloud slice in the point cloud coordinate system. Here, the three-dimensional coordinates of the preset origin point cloud corresponding to the i-th tunnel point cloud slice in the point cloud coordinate system are set as follows: Obtain the three-dimensional coordinates of each point cloud corresponding to each roadway point cloud slice in the point cloud coordinate system. Here, the three-dimensional coordinates of the j-th point cloud corresponding to the i-th roadway point cloud slice in the point cloud coordinate system are set as follows: The vectors between each point cloud and the preset origin point cloud are calculated as follows: ; in, The vector between the j-th point cloud corresponding to the i-th lane point cloud slice and the preset origin point cloud corresponding to the i-th lane point cloud slice; S3014, Determine the two-dimensional coordinates of each point cloud in the corresponding slice coordinate system based on the vector between each point cloud and the preset origin point cloud, the horizontal axis unit normal vector and the vertical axis unit normal vector; Perform a dot product operation between the vectors between each point cloud and the preset origin point cloud and the unit normal vector of the horizontal axis to determine the horizontal coordinate of each point cloud in the corresponding slice coordinate system:

[0036] in, Let be the x-coordinate of the j-th point cloud corresponding to the i-th tunnel point cloud slice in the slice coordinate system. The horizontal axis is the unit normal vector; Perform a dot product operation between the vectors between each point cloud and the preset origin point cloud and the unit normal vector along the y-axis to determine the y-coordinate of each point cloud in the corresponding slice coordinate system:

[0037] in, Let be the ordinate of the j-th point cloud corresponding to the i-th tunnel point cloud slice in the slice coordinate system. The vertical axis is the unit normal vector; the two-dimensional coordinates of each point cloud in the corresponding slice coordinate system are determined based on the horizontal and vertical coordinates of each point cloud in the corresponding slice coordinate system.

[0038] In one embodiment, this embodiment provides a specific implementation method for determining the center point coordinates of each roadway point cloud slice based on the point cloud density corresponding to each high-density grid and the centroid coordinates of each high-density grid: S3071, divide each high-density grid in each row of the point cloud slice of each roadway into a left high-density grid and a right high-density grid. Each high-density grid in each row of the point cloud slice for each roadway is divided into a left high-density grid and a right high-density grid. Based on the left and right high-density grids, the left and right walls corresponding to each roadway point cloud slice can be determined. Specifically, the left high-density grid in each row of the point cloud slice for each roadway is determined, and the q-th left high-density grid in the e-th row of the i-th roadway point cloud slice is denoted as... ; denote the r-th right-hand high-density grid in the e-th row of the i-th lane point cloud slice as Where e = 1, 2, 3, ... .

[0039] S3073, determine the centroid coordinates of the left side of each row of grids based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding left high-density grid in each row of grids; determine the centroid coordinates of the right side of each row of grids based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding right high-density grid in each row of grids. The centroid coordinates of the left side of each row of grid cells are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding high-density grid cell on the left. Specifically, since each grid cell in each tunnel point cloud slice is a square structure of the same size, the center y-coordinate of the e-th row in the i-th tunnel point cloud slice is determined by the y-coordinate of the horizontal symmetry axis of the e-th row in the coordinate system of the i-th slice. For example, if the height of each row in the i-th tunnel point cloud slice is set to... If the e-th row is located above the horizontal axis of the i-th slice coordinate system, then the value of the center ordinate of the e-th row is determined as follows:

[0040] The abscissa of the center of each high-density grid on the left in the e-th row of the i-th lane point cloud slice is determined based on the point cloud density and the abscissa of the center of the high-density grid on the left in the e-th row of the i-th lane point cloud slice. The abscissa of the center of the center of the high-density grid on the left in the e-th row of the i-th lane point cloud slice is determined by the grid width; for example, if the width of each column in the i-th lane point cloud slice is set to Then, the value of the center x-coordinate of the g-th grid starting from the vertical axis in the e-th row is: ; Calculate the product of the point cloud density of each left-hand high-density grid in the e-th row of the i-th roadway point cloud slice and the value of its central abscissa. Summate the products of the point cloud density of each left-hand high-density grid and the value of its central abscissa to obtain the first sum. Summate the point cloud densities corresponding to each left-hand high-density grid to obtain the second sum. Calculate the ratio of the first sum to the second sum to determine the central abscissa of the left-hand high-density grid in the e-th row of the i-th roadway point cloud slice. Construct the centroid coordinates of the left side of the e-th row based on the central ordinate of the e-th row and the central abscissa of the left-hand high-density grid in the e-th row. The x-coordinate of the center of the high-density grid on the right side of the e-th row in the i-th lane point cloud slice is determined based on the point cloud density and the center x-coordinate of the high-density grid on the right side of the e-th row in the i-th lane point cloud slice. The center x-coordinate of the center of the high-density grid on the right side of the e-th row in the i-th lane point cloud slice is determined by the grid width. For example, if the height of each column in the i-th lane point cloud slice is set to... Then, the value of the center x-coordinate of the G-th grid starting from the vertical axis in the e-th row is: ; Calculate the product of the point cloud density of each right-hand high-density grid in the e-th row of the i-th roadway point cloud slice and the value of the center abscissa. Summate the products of the point cloud density of each right-hand high-density grid and the value of the center abscissa to obtain the third sum. Summate the point cloud densities corresponding to each right-hand high-density grid to obtain the fourth sum. Calculate the ratio of the third sum to the fourth sum to determine the center ordinate of the right-hand high-density grid in the e-th row of the i-th roadway point cloud slice. Based on the center ordinate of the e-th row and the center abscissa of the right-hand high-density grid in the e-th row, construct the centroid coordinates of the right side of the e-th row.

[0041] S3075, Based on the left and right centroid coordinates of each row of grids in each roadway point cloud slice, determine the center point coordinates of each roadway point cloud slice. The center point coordinates of each roadway point cloud slice are determined based on the left and right centroid coordinates of each row of grids. Specifically, the average coordinates of the left and right centroid coordinates of each row of grids in each roadway point cloud slice are calculated to obtain the center point coordinates of each row. Then, the average coordinates of the center point coordinates of all rows in each roadway point cloud slice are calculated to determine the center point coordinates of each roadway point cloud slice.

[0042] In one embodiment, this embodiment provides a specific implementation method for determining the center point coordinates of each roadway point cloud slice based on the point cloud density corresponding to each high-density grid and the centroid coordinates of each high-density grid: S3072, divide each high-density grid in each column of the point cloud slice of each roadway into an upper high-density grid and a lower high-density grid. Each high-density grid in each column of the point cloud slice for each roadway is divided into upper high-density grids and lower high-density grids. Based on the upper and lower high-density grids, the upper and lower sidewalls corresponding to each roadway point cloud slice can be determined. Specifically, the upper high-density grids in each column of the point cloud slice for each roadway are determined. The Q-th upper high-density grid in the f-th row of the i-th roadway point cloud slice is denoted as... ; denote the R-th lower high-density grid in the f-th row of the i-th lane point cloud slice as Where f = 1, 2, 3, ...

[0043] S3074, determine the centroid coordinates of the upper side of each row of grids based on the point cloud density, center x-coordinate, and center y-coordinate of each upper high-density grid in each column of grids; determine the centroid coordinates of the lower side of each column of grids based on the point cloud density, center x-coordinate, and center y-coordinate of each lower high-density grid in each column of grids. Based on the point cloud density, center x-coordinate, and center y-coordinate of each column of the grid corresponding to the upper high-density grid, the centroid coordinates of the upper side of each column of the grid are determined. Specifically, since each grid in each tunnel point cloud slice is a square structure of the same size, the center x-coordinate of the f-th column in the i-th tunnel point cloud slice is determined by the x-coordinate of the y-axis of symmetry of the f-th column in the coordinate system of the i-th slice. For example, if the width of each column in the i-th tunnel point cloud slice is set to... If the f-th column is located above the vertical axis of the i-th slice coordinate system, then the value of the center x-coordinate of the f-th column is determined as follows:

[0044] The center ordinate of each upper high-density grid in the e-th column of the i-th tunnel point cloud slice is determined based on the point cloud density and center ordinate of the upper high-density grid in the f-th column of the i-th tunnel point cloud slice. The center ordinate of the upper high-density grid in the f-th column of the i-th tunnel point cloud slice is determined by the grid height. For example, if the height of each column in the i-th tunnel point cloud slice is set to... Then, the value of the center ordinate of the t-th grid in column f, starting from the horizontal axis, is: ; Calculate the product of the point cloud density of each upper high-density grid in column f of the i-th roadway point cloud slice and the value of the center ordinate. Summate the products of the point cloud density of each upper high-density grid and the value of the center ordinate to obtain the fifth sum. Summate the point cloud densities corresponding to each upper high-density grid to obtain the sixth sum. Calculate the ratio of the fifth sum to the sixth sum to determine the center ordinate of the upper high-density grid in column f of the i-th roadway point cloud slice. Construct the centroid coordinates of the upper side corresponding to column f based on the center ordinate of column f and the center abscissa of the upper high-density grid in column f. The center ordinate of each lower high-density grid in the f-th column of the i-th tunnel point cloud slice is determined based on the point cloud density and center ordinate of the lower high-density grid in the f-th column of the i-th tunnel point cloud slice. The center ordinate of the lower high-density grid in the f-th column of the i-th tunnel point cloud slice is determined by the grid height. For example, if the height of each column in the i-th tunnel point cloud slice is set to... Then, the value of the center ordinate of the T-th grid in column f, starting from the vertical axis, is: ; Calculate the product of the point cloud density of each lower high-density grid in column f of the i-th roadway point cloud slice and the value of the center ordinate. Summate the products of the point cloud density of each lower high-density grid and the value of the center ordinate to obtain the seventh sum. Summate the point cloud densities corresponding to each lower high-density grid to obtain the eighth sum. Calculate the ratio of the seventh sum to the eighth sum to determine the center ordinate of the lower high-density grid in column f of the i-th roadway point cloud slice. Construct the centroid coordinates of the lower side corresponding to column f based on the center abscissa of column f and the center ordinate of the lower high-density grid in column f.

[0045] S3076, Determine the center point coordinates of each roadway point cloud slice based on the upper and lower centroid coordinates of each column of grid corresponding to each roadway point cloud slice; The center point coordinates of each roadway point cloud slice are determined based on the upper and lower centroid coordinates of each column of grid in each roadway point cloud slice. Specifically, the average coordinates of the upper and lower centroid coordinates of each column of grid in each roadway point cloud slice are calculated to obtain the center point coordinates of each column. Then, the average coordinates of the center point coordinates of all columns in each roadway point cloud slice are calculated to determine the center point coordinates of each roadway point cloud slice.

[0046] In one embodiment, this embodiment provides a specific implementation method for obtaining the centerline of the roadway to be calibrated by performing curve fitting on the center points corresponding to each roadway point cloud slice based on the coordinates of each center point: S401, transform the center point coordinates corresponding to each roadway point cloud slice from the slice coordinate system to the point cloud coordinate system; S403, based on the curve fitting method and the coordinates of each center point in the point cloud coordinate system, curve fitting is performed on the center points corresponding to each roadway point cloud slice to obtain the roadway centerline to be calibrated; wherein, the curve fitting method includes the B-spline curve fitting method. The coordinates of the center points corresponding to each roadway point cloud slice are transformed from the slice coordinate system to the point cloud coordinate system. Specifically, the horizontal displacement coordinate is obtained by cross-product of the horizontal coordinate and the horizontal axis unit normal vector of the center point coordinates of each roadway point cloud slice. The vertical displacement coordinate is obtained by cross-product of the vertical coordinate and the vertical axis unit normal vector of the center point coordinates of each roadway point cloud slice. The horizontal displacement coordinate and the vertical displacement coordinate are summed to obtain the total displacement coordinate. The three-dimensional coordinates of the preset origin point cloud corresponding to each roadway point cloud slice in the point cloud coordinate system are determined. The three-dimensional coordinates of each preset origin point cloud in the point cloud coordinate system are summed with the total displacement coordinates of the center point to obtain the three-dimensional coordinates of the center point in the point cloud coordinate system. Finally, based on the curve fitting method and the coordinates of each center point in the point cloud coordinate system, the center line of the roadway to be calibrated is obtained by curve fitting of the center point corresponding to each roadway point cloud slice.

[0047] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0048] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0049] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0050] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, 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, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for obtaining the centerline of a roadway, characterized in that, include: Obtain the point cloud model of the tunnel to be calibrated; wherein the point cloud model of the tunnel is located in the point cloud coordinate system; Multiple lane point cloud slices distributed along the principal component direction of the lane point cloud model are generated based on the point cloud slice spacing and point cloud slice thickness. Each of the roadway point cloud slices is divided into multiple grids. The grids in each roadway point cloud slice where the point cloud density is greater than a preset density threshold are taken as high-density grids. The center point coordinates of each roadway point cloud slice are determined based on the centroid coordinates of each high-density grid. Based on the coordinates of each center point, curve fitting is performed on the center points corresponding to each tunnel point cloud slice to obtain the tunnel centerline of the tunnel to be calibrated.

2. The method for obtaining the centerline of a roadway according to claim 1, characterized in that, The step of generating multiple tunnel point cloud slices distributed along the principal component direction of the tunnel point cloud model based on the point cloud slice spacing and point cloud slice thickness includes: The point clouds in the tunnel point cloud model are preprocessed to obtain the preprocessed tunnel point cloud model; wherein, the preprocessing includes: denoising processing and downsampling processing; Principal component analysis is performed on the preprocessed tunnel point cloud model to determine the principal component orientation of the tunnel point cloud model; Based on the point cloud slice spacing and the point cloud slice thickness, multiple lane point cloud slices are generated that are distributed along the principal component direction of the lane point cloud model.

3. The method for obtaining the centerline of a roadway according to claim 2, characterized in that, The step of preprocessing each point cloud in the lane point cloud model includes: The roadway point cloud model is denoised using the statistical outlier method to remove outlier point clouds from the model, resulting in a denoised roadway point cloud model. The denoised tunnel point cloud model is downsampled using a voxel grid filter with a preset voxel size to obtain the preprocessed tunnel point cloud model.

4. The method for obtaining the centerline of a roadway according to claim 2, characterized in that, The step of performing principal component analysis on the preprocessed tunnel point cloud model to determine the principal component orientation of the tunnel point cloud model includes: Obtain the centroid coordinates of the preprocessed tunnel point cloud model, center the coordinates of each point cloud in the tunnel point cloud model based on the centroid coordinates, and construct the point cloud matrix corresponding to the tunnel point cloud model based on the coordinates of each point cloud after centering. Based on the point cloud matrix, the covariance matrix corresponding to the tunnel point cloud model is determined, and multiple eigenvalues ​​are obtained by performing feature analysis on the covariance matrix. The direction represented by the eigenvector corresponding to the largest eigenvalue is taken as the principal component direction of the tunnel point cloud model.

5. The method for obtaining the centerline of a roadway according to claim 1, characterized in that, The step of dividing each of the roadway point cloud slices into multiple grids, designating the grids in each roadway point cloud slice where the point cloud density is greater than a preset density threshold as high-density grids, and determining the center point coordinates of each roadway point cloud slice based on the centroid coordinates of each high-density grid includes: Establish a slice coordinate system corresponding to each of the roadway point cloud slices, and transform the coordinates of the point cloud contained in each of the point cloud slices from the point cloud coordinate system to the slice coordinate system; Each of the aforementioned tunnel point cloud slices is divided into multiple grids, and the point cloud density corresponding to each grid is determined based on the number of point clouds contained in each grid. Determine the high-density grid corresponding to the point cloud density that is greater than a preset density threshold in each of the roadway point cloud slices; The center point coordinates of each roadway point cloud slice are determined based on the point cloud density corresponding to each high-density grid in each roadway point cloud slice and the centroid coordinates of each high-density grid; wherein the centroid coordinates of the high-density grid and the center point coordinates are located in the slice coordinate system.

6. The method for obtaining the centerline of a roadway according to claim 5, characterized in that, The step of establishing the slice coordinate system corresponding to each of the tunnel point cloud slices includes: The orientation unit normal vector of the roadway point cloud model is determined based on the three-dimensional coordinates of the preset starting point cloud and the preset ending point cloud in the point cloud coordinate system; wherein, the direction of the orientation unit normal vector is the same as the direction of the principal component. Obtain the corresponding gravity unit normal vector in the world coordinate system, and determine the vertical axis unit normal vector in the slice coordinate system corresponding to each of the tunnel point cloud slices based on the orientation unit normal vector and the gravity unit normal vector; Based on the directional unit normal vector and the vertical axis unit normal vector, determine the horizontal axis unit normal vector in the slice coordinate system corresponding to each of the roadway point cloud slices; Based on the preset origin point cloud in each of the tunnel point cloud slices, the direction of the horizontal axis unit normal vector, and the direction of the vertical axis unit normal vector, a slice coordinate system corresponding to each of the tunnel point cloud slices is constructed.

7. The method for obtaining the centerline of a roadway according to claim 6, characterized in that, The step of transforming the coordinates of the point cloud contained in each point cloud slice from the point cloud coordinate system to the slice coordinate system includes: The vector between each point cloud and the preset origin point cloud is determined based on the preset origin point cloud corresponding to each of the tunnel point cloud slices and the three-dimensional coordinates of each point cloud in the point cloud coordinate system; The two-dimensional coordinates of each point cloud in the corresponding slice coordinate system are determined based on the vector between each point cloud and the preset origin point cloud, the horizontal axis unit normal vector, and the vertical axis unit normal vector.

8. The method for obtaining the centerline of a roadway according to claim 5, characterized in that, The step of determining the center point coordinates of each roadway point cloud slice based on the point cloud density of each high-density grid and the centroid coordinates of each high-density grid in each roadway point cloud slice includes: Each high-density grid in each row of the grid of each tunnel point cloud slice is divided into a left high-density grid and a right high-density grid. The centroid coordinates of the left side of each row of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding left high-density grid in each row of grids; the centroid coordinates of the right side of each row of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding right high-density grid in each row of grids. The center point coordinates of each tunnel point cloud slice are determined based on the left and right centroid coordinates of each row of grids.

9. The method for obtaining the centerline of a roadway according to claim 5, characterized in that, The step of determining the center point coordinates of each roadway point cloud slice based on the point cloud density of each high-density grid and the centroid coordinates of each high-density grid in each roadway point cloud slice includes: Each high-density grid in each column of the grid of each of the aforementioned roadway point cloud slices is divided into an upper high-density grid and a lower high-density grid; The centroid coordinates of the upper side of each row of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding upper high-density grid in each column of grids; the centroid coordinates of the lower side of each column of grids are determined based on the point cloud density, center x-coordinate, and center y-coordinate of each corresponding lower high-density grid in each column of grids. The center point coordinates of each tunnel point cloud slice are determined based on the upper and lower centroid coordinates of each column of the grid.

10. The method for obtaining the centerline of a roadway according to claim 5, characterized in that, The step of performing curve fitting on the center points corresponding to the point cloud slices of each roadway based on the coordinates of each center point to obtain the roadway centerline of the roadway to be calibrated includes: The coordinates of the center point corresponding to each of the roadway point cloud slices are transformed from the slice coordinate system to the point cloud coordinate system; Based on the curve fitting method and the coordinates of each center point in the point cloud coordinate system, the center point corresponding to each roadway point cloud slice is curve fitted to obtain the roadway centerline to be calibrated; wherein, the curve fitting method includes the B-spline curve fitting method.

Images