Extraction system and method for laser point cloud vertical and horizontal section data

Abstract

The invention provides an extraction system and method for laser point cloud longitudinal and transverse section data, and relates to the technical field of surveying and mapping engineering and computer aided design, and the system comprises a normal production module which is used for analyzing midline data, and generating a first normal file with coordinate information and a second normal file with attribute information; the point cloud processing module is used for constructing a ground surface model, projecting the first normal file to the ground surface model to calculate three-dimensional coordinate information of the normal file, and outputting a third normal file with the three-dimensional coordinate information; and the section production module is used for generating section data and longitudinal section data in preset formats according to the second normal file and the third normal file. The system has high flexibility on the premise of ensuring data precision, is suitable for design, construction and monitoring of linear projects such as roads, railways and pipelines and spatial analysis and decision support in a geographic information system, and can realize efficient and accurate extraction and application of laser point cloud data.

Description

Technical Field

[0001] This application relates to the technical fields of surveying and mapping engineering and computer-aided design, and particularly relates to a system and method for extracting longitudinal and cross-sectional data of laser point clouds. Background Art

[0002] In the fields of engineering surveying, road design, and geological exploration, the generation of longitudinal and cross-sectional data is a fundamental and crucial link for carrying out terrain analysis and engineering design. Currently, traditional cross-section production methods mainly rely on the following two types of technical solutions, but in the actual application process, significant limitations have emerged.

[0003] I. Manual acquisition method:

[0004] Currently, some projects still use the full-field GPS measurement point method to obtain cross-section data. The specific operation is for manual carrying of measurement equipment to collect elevation and coordinate information point by point in the target area. Although this method can ensure the accuracy of data to a certain extent, its existing defects cannot be ignored:

[0005] High labor and time costs: It is necessary to organize a professional measurement team to carry out large-scale field operations. Especially in the face of complex terrain or large-scale survey scenarios, the data collection cycle is long and the labor input is extremely large.

[0006] Poor environmental adaptability: It is easily restricted by factors such as weather conditions and terrain obstacles. It is difficult to achieve data collection coverage in some areas such as steep slopes and dense forests, resulting in poor data integrity.

[0007] Complicated post-processing: The original measurement point data needs to go through multiple steps such as manual sorting, coordinate conversion, and interpolation calculation. It is not only easy to introduce human errors, but also the data processing efficiency is very low.

[0008] II. Multi-software collaborative processing method:

[0009] In order to improve work efficiency, multiple professional software, such as ArcGIS, Terrasolid, AutoCAD, etc., are often used for collaborative operations in the prior art. Although this method can improve efficiency to a certain extent, there are still many problems:

[0010] Operational fragmentation: It is necessary to frequently switch between different software interfaces, and information loss or compatibility errors are likely to occur during the data format conversion process. Any small mistake may lead to rework of the entire process, greatly increasing the processing difficulty and time cost.

[0011] Low degree of automation: Key steps, such as node matching, distance and height difference calculation, etc., rely on manual intervention, and repetitive operations are prone to errors.

[0012] In summary, in terms of efficiency, accuracy, and automation, the above methods are difficult to meet the requirements of modern engineering for rapid and intelligent cross-section production. Therefore, it is necessary to develop a cross-section extraction method based on laser point clouds to solve the above technical problems. Summary of the Invention

[0013] This application provides a system and method for extracting laser point cloud longitudinal and cross-section data.

[0014] In a first aspect, an embodiment of this application provides a system for extracting laser point cloud longitudinal and cross-section data, including:

[0015] A normal line production module for parsing center line data to generate a first normal line file with coordinate information and a second normal line file with attribute information;

[0016] A point cloud processing module for constructing a surface model, projecting the first normal line file onto the surface model to calculate the three-dimensional coordinate information of the first normal line file, and outputting a third normal line file with the three-dimensional coordinate information;

[0017] A cross-section production module for generating one or more of cross-section data and longitudinal section data in a preset format according to the second normal line file and the third normal line file.

[0018] In a second aspect, an embodiment of this application provides a method for extracting laser point cloud longitudinal and cross-section data, which may include:

[0019] Parsing center line data to generate a first normal line file with coordinate information and a second normal line file with attribute information;

[0020] Constructing a surface model, projecting the first normal line file onto the surface model to calculate the three-dimensional coordinate information of the first normal line file, and outputting a third normal line file with the three-dimensional coordinate information;

[0021] Generating one or more of cross-section data and longitudinal section data in a preset format according to the second normal line file and the third normal line file.

[0022] In some embodiments, the method of parsing center line data includes:

[0023] Extracting geometric parameters and calculating the line segment length;

[0024] Connecting and integrating the discrete line segments in the center line data;

[0025] Determining the direction of each line segment in the center line data;

[0026] Calculate the total length, offset, and normal projection points of continuous line segments;

[0027] Determine the normal direction and calculate the endpoint coordinates of the normal.

[0028] In some embodiments, extracting geometric parameters and calculating the line segment length includes:

[0029] For a line segment entity or a polyline entity, extract the endpoint coordinates and calculate the line segment length or polyline length; for an arc entity or a circle entity, extract the center, the radius, the start and end angles, calculate the arc length and the discrete point coordinates to extract geometric parameters and calculate the line segment length;

[0030] The connection integrates the discrete line segments in the center line data, including:

[0031] Construct a set of all line segment endpoints And set a tolerance threshold ε. If the endpoints of two lines satisfy Then it is determined as a continuous line segment, and the adjacent line segments with common endpoints are connected; for a non - continuous path, connect the current path end point to the candidate line segment start point to connect and integrate the discrete line segments in the center line data;

[0032] The determination of the direction of each line segment in the center line data includes:

[0033] Receive the start point direction parameter, and judge the coordinate relationship between the initial endpoint P start Of the continuous line segment and the termination endpoint P end According to the coordinate relationship, determine the start point and end point of each line segment, and normalize the direction of the center line from the start point to the end point to determine the direction of each line segment in the center line data;

[0034] The calculation of the total length, station offset, and normal projection points of continuous line segments includes:

[0035] Calculate the Euclidean distance between continuous points, and determine the total length of the continuous line segment based on the Euclidean distance;

[0036] Read the station number from the engineering document and generate a sequence of station number offsets, and determine the complete set of station numbers S based on the sequence of station number offsets;

[0037] For each station number s i ∈S, calculate the cumulative path length s acc =s i -s0; then traverse the geometric path sub - segments{(p k ,p k+1 ,L k )}, locate the target point by piece - wise accumulation. When s prev ≤sacc ≤s prev +L k In the case of, calculate the local ratio where s prev is the previous cumulative length, o k is the starting point of the sub-segment, p k+1 is the ending point of the sub-segment, L k is the Euclidean distance of the sub-segment; calculate the interpolation coordinates of the target point as (x, y) = (x k + λ(x k+1 - x k ), y k + λ(y k+1 - y k ))), calculate the tangent vector of the current sub-segment and determine the normal projection point after normalizing the tangent vector;

[0038] The determining the normal direction and calculating the endpoint coordinates of the normal includes:

[0039] Determine the left normal and the right normal according to the normalized tangent vector, and determine the endpoint coordinates of the normal according to the preset left and right buffer distances.

[0040] In some embodiments, the constructing the surface model includes:

[0041] Construct a Delaunay triangulation based on the input laser point cloud ground point data, and construct the surface model based on the Delaunay triangulation.

[0042] In some embodiments, the generating one or more of the cross-sectional data and the longitudinal-sectional data in a preset format according to the second normal file and the third normal file includes:

[0043] Extract the 3D polyline and the 3D coordinates of the nodes on the polyline from the second normal file and the third normal file;

[0044] Fuse the attribute information of the input data with the extracted 3D coordinates of the nodes to attach the corresponding attribute information to the 3D coordinates of the nodes;

[0045] For the nodes of the same normal, sort them in ascending order according to the distance of the nodes from the midline projection;

[0046] Calculate the planar distance and the height difference between adjacent nodes;

[0047] Set a filtering threshold, and thin out the nodes according to the filtering threshold, the Euclidean distance, and the height difference.

[0048] In some embodiments, fusing the attribute information of the input data with the extracted three-dimensional coordinates of the nodes to attach corresponding attribute information to the three-dimensional coordinates of the nodes includes:

[0049] Based on the matching relationship of spatial positions, performing an intersection operation on the planar coordinates of the nodes and the geospatial information in the second normal file to obtain the three-dimensional coordinates of the nodes with corresponding attribute information attached;

[0050] Setting the filtering threshold, and thinning the nodes according to the filtering threshold, the Euclidean distance, and the elevation difference, includes:

[0051] Traversing the node sequence. When the planar distance D between adjacent nodes i is less than the minimum distance threshold ε, marking the two nodes as coincident points and merging them; when the elevation difference threshold ΔH i is greater than the first elevation difference threshold, sending a manual review prompt message; when the elevation difference threshold ΔH i is less than the second elevation difference threshold, deleting the node sequence with an elevation difference less than the elevation difference threshold.

[0052] In some embodiments, calculating the planar distance and elevation difference between adjacent nodes includes:

[0053] For adjacent nodes p i =(x i , y i , z i ) and p i+1 =(x i+1 , y i+1 , z i+1 ), its planar distance is calculated according to the Euclidean distance, and the Euclidean distance

[0054] Calculating the vertical direction difference ΔH between adjacent nodes based on the three-dimensional coordinates i =Z i+1 -Z i .

[0055] In some embodiments, the file header identifier of the longitudinal section data in the preset format is HINTCAD5.83_DMX_SHUJU; each line records the station number and its reference elevation. The file header identifier of the cross-section data in the preset format is HINTCAD5.83_HDM_SHUJU; grouped by station number, each station number data block contains: station number value, number of left / right section points, alternating planar distance changes and elevation differences.

[0056] Compared with the prior art, the beneficial effects of the present application are as follows: By parsing the centerline data, a first normal file with coordinate information and a second normal file with attribute information are generated, laying a foundation for subsequent processing. Subsequently, a surface model is constructed, and the first normal file is projected into the surface model to calculate three-dimensional coordinate information, and a third normal file containing three-dimensional coordinates is output to ensure the spatial accuracy of the data. Finally, by combining the second normal file and the third normal file, one or more of the cross-sectional data and longitudinal-sectional data in a preset format are generated to meet different engineering requirements. This method improves efficiency through automated processing, ensures data accuracy by combining the surface model and three-dimensional coordinate calculation, and has high flexibility. It is applicable to the design, construction, and monitoring of linear projects such as roads, railways, and pipelines, as well as spatial analysis and decision support in geographic information systems (GIS), realizing the efficient and accurate extraction and application of laser point cloud data. Description of the Drawings

[0057] Figure 1 It is a schematic structural diagram of an extraction system for laser point cloud cross-sectional and longitudinal-sectional data provided by an embodiment of the present application.

[0058] Figure 2 It is a schematic flow diagram of an extraction system for laser point cloud cross-sectional and longitudinal-sectional data provided by an embodiment of the present application. Detailed Embodiments

[0059] The present application will be further described in detail below in conjunction with test examples and specific embodiments. However, it should not be understood that the scope of the above-mentioned subject matter of the present application is limited to the following embodiments. Any technology implemented based on the content of the present application belongs to the scope of protection of the present application.

[0060] In the description of the specific embodiments of the present application, without special explanation, the expression terms indicating the orientation or positional relationship such as "upper", "lower", "left", "right", "center", "inner", "outer", "side", etc. are all based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship when the product / device / device is usually used and placed. These orientation or positional relationship terms are only for the convenience of describing the solution of the present application or simplifying the description in the specific embodiment, facilitating technicians to quickly understand the solution, rather than indicating or implying that a specific device / component / element must have a specific orientation or be constructed and operated in a specific positional relationship. Therefore, it should not be construed as a limitation to the present application.

[0061] In the description of the embodiments of the present application, technical terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity, specific order, or primary-secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" or "multiple types" is two or more, unless otherwise clearly and specifically defined.

[0062] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0063] Embodiment 1

[0064] Please refer to Figure 1 and Figure 2 , Figure 1 which is a schematic structural diagram of the extraction system for laser point cloud longitudinal and cross-sectional data provided by the embodiments of the present application. Figure 2 which is a schematic flow diagram of the extraction system for laser point cloud longitudinal and cross-sectional data. The extraction system 10 for laser point cloud longitudinal and cross-sectional data may include:

[0065] A normal production module 11 for parsing center line data to generate a first normal file with coordinate information and a second normal file with attribute information.

[0066] Among them, the normal production module 11 can use relevant libraries of Python to read the DXF center line file, parse and calculate geometric parameters, connect discrete line segments, determine the center line direction and length, calculate the offset, normal projection point, normal direction, and endpoint coordinates, and finally generate a first normal file, that is, the normal.DXF file, and a second normal file, that is, the normal.SHP file. The normal production module 11 can use the ezdxf library of Python to read the DXF file and traverse the geometric entities in the model space (Modelspace), including lines (Line), polylines (LWPOLYLINE, POLYLINE), arcs (ARC), circles (CIRCLE), etc.

[0067] A point cloud processing module 12 for constructing a surface model, projecting the first normal file onto the surface model to calculate the three-dimensional coordinate information of the first normal file, and outputting a third normal file with the three-dimensional coordinate information.

[0068] The point cloud processing module 12 can import the laser point cloud ground point data and the normal.DXF data in the Terrasolid software, construct a Delaunay triangulation network, project the normal to the surface model to obtain three-dimensional coordinate information, and finally export the third normal file, that is, the data normal (attribute).DXF.

[0069] The cross-section production module 13 is used to generate one or more of the cross-section data and the longitudinal section data in a preset format according to the second normal file and the third normal file.

[0070] The cross-section production module 13 can use the Python library to read relevant files, extract node coordinates and assign normal attributes, sort the node sequence, calculate the plane distance and the height difference, filter the nodes by a threshold, and finally output the longitudinal section.dmx and the cross-section.hdm files in the Hintcad format. If the normal (attribute).DXF is divided into several segments of data for separate processing, the processing time will be reduced accordingly by a multiple.

[0071] Embodiment 2

[0072] This embodiment is an example of applying the extraction system for the laser point cloud longitudinal and cross-section data in the above Embodiment 1 to the extraction method of the laser point cloud longitudinal and cross-section data. The method provided by this application can be deeply integrated based on Python and Terrasolid to realize the full-process rapid generation of the laser point cloud data into the longitudinal and cross-section data. The steps of the extraction method for the laser point cloud longitudinal and cross-section data may include:

[0073] S1. Analyze the center line data to generate a first normal file with coordinate information and a second normal file with attribute information.

[0074] S2. Construct a surface model, project the first normal file to the surface model to calculate the three-dimensional coordinate information of the first normal file, and output a third normal file with the three-dimensional coordinate information.

[0075] S3. Generate one or more of the cross-section data and the longitudinal section data in a preset format according to the second normal file and the third normal file.

[0076] In the above implementation process, the first normal file with coordinate information and the second normal file with attribute information are generated by parsing the centerline data, laying a foundation for subsequent processing; then a surface model is constructed, and the first normal file is projected into the surface model to calculate three-dimensional coordinate information, and a third normal file containing three-dimensional coordinates is output to ensure the spatial accuracy of the data; finally, by combining the second normal file and the third normal file, one or more of the cross-sectional data and longitudinal-sectional data in a preset format are generated to meet different engineering requirements. This method improves efficiency through automated processing, ensures data accuracy by combining the surface model and three-dimensional coordinate calculation, and has high flexibility. It is applicable to the design, construction, and monitoring of linear projects such as roads, railways, and pipelines, as well as spatial analysis and decision support in geographic information systems (GIS), realizing the efficient and accurate extraction and application of lidar point cloud data.

[0077] Embodiment 3

[0078] This embodiment is an example of parsing the centerline data in the above Embodiment 2.

[0079] The ways of parsing the centerline data may include:

[0080] Extracting geometric parameters and calculating the line segment length: For a line segment entity or a polyline entity, extract the endpoint coordinates and calculate the line segment length or polyline length; for an arc entity or a circle entity, extract the center, the radius, the start and end angles, calculate the arc length and the discrete point coordinates to extract geometric parameters and calculate the line segment length. Specifically, the shapely library of Python can be used to calculate the key geometric information for each extracted geometric element. For a straight line and a polyline, the Euclidean distance is calculated through its endpoint coordinates (x1, y1) and (x2, y2) to obtain the line segment length. For an arc and a circle, according to its center (x c , y c ), radius r, the central angle Δθ converted to radians, the number of discrete points and other parameters, calculate the arc length L = r·Δθ and the coordinates of any point on the arc (x i , y i ) = (x c + rcosθ i , y c + rsinθ i ); the parameter extraction methods for a polyline and a line segment are the same, and the parameter extraction methods for a circle and an arc are the same.

[0081] Connecting and integrating the discrete line segments in the centerline data: Construct a set of all line segment endpoints and set a tolerance threshold ε. If the endpoints of two lines satisfy It is determined as a continuous line segment, and adjacent line segments with a common endpoint are connected. For a discontinuous path, the end point of the current path is connected to the start point of the candidate line segment to connect and integrate the discrete line segments in the center line data. Exemplarily, the tolerance threshold ε can default to 0.01 m.

[0082] Determine the direction of each line segment in the center line data: Receive the start direction parameter, and judge the initial endpoint P of the continuous line segment according to the start direction parameter start and the termination endpoint P end 's coordinate relationship. Determine the start and end points of each line segment according to the coordinate relationship, and normalize the direction of the center line from the start point to the end point to determine the direction of each line segment in the center line data. Exemplarily, according to the start direction parameter specified by the user (1: upper end / 2: lower end / 3: left end / 4: right end), judge the continuous line segment P start and the termination endpoint P end 's coordinate relationship: If the direction parameter is specified as 1, that is, the y coordinate of Pstart > the y coordinate of Pend, then the upper end is the start point; if the direction parameter is specified as 2, that is, the y coordinate of Pstart < the y coordinate of Pend, then the lower end is the start point; if the direction parameter is specified as 3, that is, the x coordinate of Pstart > the x coordinate of Pend, then the left end is the start point; if the direction parameter is specified as 4, that is, the x coordinate of Pstart < the x coordinate of Pend, then the right end is the start point. Normalize the center line direction from the start point to the end point.

[0083] Calculate the total length, offset, and normal projection point of the continuous line segment: For the total length of the center line, calculate the Euclidean distance between consecutive points, and determine the total length of the continuous line segment based on the Euclidean distance. For example: For the integrated continuous point sequence {p0, p1,..., pn}, calculate the Euclidean distance segment by segment For isolated line segments that cannot be connected, retain the original endpoint information and output a warning log; when the deviation between the total path length and the sum of the discrete line segment lengths exceeds 2ε, trigger the fault tolerance mechanism to re - perform the segmentation calculation.

[0084] For the treatment of the center stake number offset, the stake number can be read from the engineering file and a sequence of stake number offsets is generated. Determine the complete set of stake numbers S based on the sequence of stake number offsets. For example: Read the stake number from an external file and convert it into a sequence of stake number offsets {Δs1, Δs2,..., Δs n}, and calculate the actual stake number value in combination with the starting stake number s0: s i = s0 + Δs i (i = 1, 2,..., n), and then automatically add the stake number s corresponding to the total path length end = s0 + L total , to form the complete set of stake numbers S = {s o , s1,..., s n,s end}。

[0085] For the calculation of the normal projection point, for each station number s i ∈S, calculate the cumulative path length s acc = s i - s0; then traverse the geometric path sub-segments {(p k , p k+1 , L k )}, locate the target point by piecewise accumulation. When s prev ≤ s acc ≤ s prev + L k is satisfied, calculate the local ratio where s prev is the previous cumulative length, p k is the starting point of the sub-segment, p k+1 is the ending point of the sub-segment, and L k is the Euclidean distance of the sub-segment; calculate the interpolation coordinates of the target point as (x, y) = (x k + λ(x k+1 - x k ), y k + λ(yk + 1 - yk), calculate the tangent vector t = (xk + 1 - xk, yk + 1 - yk) of the current sub-segment, and determine the normal projection point after normalizing the tangent vector.

[0086] For determining the normal direction and calculating the endpoint coordinates of the normal: Determine the left normal and the right normal according to the normalized tangent vector, and determine the endpoint coordinates of the normal according to the preset left and right buffer distances. For example: According to the tangent direction after normalization calculate the left normal and the right normal and calculate the left and right normal endpoint coordinates respectively according to the set left and right buffer distances B L and B R , as follows:

[0087] Example 4

[0088] This example is an example of constructing a surface model in Example 2.

[0089] Constructing a surface model may include:

[0090] Construct a Delaunay Triangulation based on the input laser point cloud ground point data, and construct the surface model based on the Delaunay Triangulation. Exemplarily, the laser point cloud data can be first imported into point cloud processing software, and the point cloud is subjected to spatial filtering and denoising processing to retain the ground classification point set P ground ={(x i ,y i ,z i )}; Then construct a Delaunay Triangulation based on the three-dimensional coordinates. The triangles of this triangulation are distributed as equilateral as possible and do not contain other points except the three vertices of the triangle; Finally, perform topological optimization on the triangulation, remove the hanging triangles and fill the holes to generate a continuous surface model T surface . After constructing the surface model, load the normal DXF file, and perform a surface projection algorithm on the two-dimensional coordinates (x i ,y i ) of the nodes thereon. Locate the triangle △ABC containing this point in T surface , and calculate the elevation value z i through barycentric coordinate interpolation. The formula is:

[0091]

[0092]

[0093] λ C =1 - λ A - λ B

[0094] z i =λ A z A + λ B z B + λ C z C

[0095] where λ A , λ B and λ C are the barycentric coordinate weights of the node (x i ,y i ) relative to the triangle △ABC respectively.

[0096] Example 5

[0097] This example is an example of generating cross-sectional data and longitudinal-sectional data in a preset format in Example 2.

[0098] The generating one or more of the cross-sectional data and the longitudinal-sectional data in the preset format according to the second normal file and the third normal file may include:

[0099] Extract three-dimensional polylines and the three-dimensional coordinates of the nodes on the polylines from the second normal file and the third normal file. For example, objects with geometric types such as lines and polylines can be traversed one by one to accurately extract the coordinate information of all their nodes, and the extracted coordinate information can be added to a pre-set coordinate group in an orderly manner to ensure the integrity and systematicness of the coordinate information, providing comprehensive and accurate basic data for subsequent operations such as attribute fusion and node sorting.

[0100] Fuse the attribute information of the input data with the extracted three-dimensional coordinates of the nodes, so that the three-dimensional coordinates of the nodes are attached with corresponding attribute information. For example, operations such as reading data containing three-dimensional coordinates of nodes and normal SHP files can be performed. Using professional data parsing techniques, station number and azimuth attribute information can be accurately extracted from the read data. Next, using spatial analysis methods, an intersection operation is performed between the planar coordinates of the nodes and the geospatial information in the SHP file. In this process, based on the matching relationship of spatial positions, attribute information such as station number and azimuth is associated and integrated with the three-dimensional coordinates of the nodes. Finally, a point file with complete attribute information such as three-dimensional coordinates, station number, and azimuth is generated, providing a more comprehensive and detailed data basis for subsequent data processing and analysis.

[0101] For the nodes of the same normal, sort them in ascending order according to the distance of the nodes from the projection of the median line. For example, sorting can be performed according to the station number and the left and right normal point sets can be grouped according to the azimuth. and Then sort each group of nodes in ascending order according to the projection distance to ensure that p i and p i+1 are continuous along the normal direction.

[0102] Calculate the planar distance and height difference between adjacent nodes. For example, for planar distance calculation, for adjacent nodes p i =(x i , y i , z i ) and p i+1 =(x i+1 , y i+1 , z i+1 ), its planar distance is calculated according to the Euclidean distance, and the Euclidean distance For height difference calculation, the vertical direction difference ΔH between adjacent nodes can be calculated based on the three-dimensional coordinates i =Z i+1 -Z i .

[0103] Set a filtering threshold, and perform thinning processing on the nodes according to the filtering threshold, the Euclidean distance, and the height difference. For example, the node sequence can be traversed. When the planar distance D between adjacent nodesi When it is less than the minimum distance threshold ε, mark the two nodes as coincident points and perform a merging process; at the height difference threshold ΔH i When it is greater than the first height difference threshold, send a manual review prompt message; at the height difference threshold ΔH i When it is less than the second height difference threshold, delete the node sequence that is less than the height difference threshold. Among them, the first height difference threshold can be 100m, and the second height difference threshold can be 0.01m. This threshold is for thinning nodes at flat ground areas and reducing output redundancy.

[0104] Finally, output the file header identifier HINTCAD5.83_DMX_SHUJU of the longitudinal section data in a preset format; each line records the station number and its reference elevation, and output the file header identifier HINTCAD5.83_HDM_SHUJU of the cross-section data in a preset format; group by station number, and each station number data block contains: the station number value, the number of left / right section points, the alternating arrangement of the plane distance change and the height difference.

[0105] Example 6

[0106] This example takes a survey project of a section of highway reconstruction and expansion as an example to illustrate the application of the method of this application to generate longitudinal and cross-section data.

[0107] I. Input data:

[0108] 1. Center line file: Provide a center line file in the DXF format. The total length of the highway design center line recorded in the center line file is 4100.67m. This file accurately records the geometric shape of the highway design center line, including various geometric elements such as line segments and arcs, providing a basic framework for subsequent normal line calculation and cross-section generation.

[0109] 2. Intermediate pile file: The intermediate pile file is in the TXT format and contains 412 station number offset data. These data are station number offset information set uniformly on the center line or according to the terrain change characteristics based on the engineering design requirements, and are used to determine the position of the normal line on the center line and calculate the station number.

[0110] 3. Laser point cloud data: The laser point cloud data file format is LAS and contains 2.7 million ground point data. These data are obtained through laser scanning technology and accurately reflect the terrain surface information of the project area, providing rich data support for constructing the surface model and obtaining the normal line elevation.

[0111] 4. Parameter settings: including the left and right normal lengths, starting stake number, and height difference threshold. According to the engineering design standards, the left and right normal lengths are both set to 20 m to ensure that the cross-section data obtained can cover the range required for highway design and meet the engineering design's requirements for terrain information; the starting stake number is set to the actual starting stake number 0; the height difference threshold is set to 0.01, that is, if the height difference between adjacent points is less than 0.01 m, the point is skipped.

[0112] II. Processing flow:

[0113] 1. The normal production module uses relevant Python libraries to read the center line file in DXF, parse and calculate geometric parameters, connect discrete line segments, determine the center line direction and length, calculate the offset, normal projection point, normal direction, and endpoint coordinates, and finally generate the normal.DXF and normal.SHP files.

[0114] 2. The point cloud processing module imports the laser point cloud ground point data and normal.DXF data in Terrasolid software, constructs a Delaunay triangulation network, projects the normal to the surface model to obtain three-dimensional coordinate information, and finally exports the data normal (attribute).DXF.

[0115] 3. The cross-section production module uses Python libraries to read relevant files, extract node coordinates and assign normal attributes, sort the node sequence, calculate the planar distance and height difference, filter nodes by the threshold, and finally output the longitudinal section.dmx and cross-section.hdm files in the Hintcad format. If the normal (attribute).DXF is divided into several segments of data for separate processing, the processing time will be reduced accordingly by multiples.

[0116] III. Result verification:

[0117] 1. Efficiency comparison: Compared with the traditional manual collection method, the processing efficiency of the present invention is greatly improved. The traditional manual collection method requires organizing a professional survey team for field operations, with a long data collection cycle and cumbersome post-processing. In this project, the traditional method takes about 10 days to complete data collection and processing of the same scale, while the present invention only takes about 14 minutes, with a significant improvement in efficiency. Compared with the multi-software collaborative processing method, the present invention avoids frequent software switching and manual intervention, and the processing time is shortened from the original about 30 minutes to 14 minutes, with the efficiency increased by 2 times.

[0118] 2. Accuracy verification: By comparing with the data generated by the multi-software collaborative processing method, the accuracy of the longitudinal and cross-section data generated by the present invention is verified. Among the 100 key points selected, the elevation error between the data generated by the present invention and the data generated by the multi-software collaborative processing method is less than 0.02 m, meeting the requirements of the engineering design for the accuracy of terrain data. However, the present invention has more advantages in terms of the stability and reliability of data processing.

[0119] Compared with the prior art, the beneficial effects of the present invention include:

[0120] (1) High efficiency:

[0121] The present invention realizes the full-process rapid processing from DXF file parsing to cross-section data output. Compared with the manual acquisition method, it does not require a large amount of manpower to participate in field operations and cumbersome post-processing of data, saving a large amount of time and labor costs; compared with the multi-software collaborative processing method, it avoids frequent software switching and manual intervention, improving the processing efficiency. Through actual tests, for processing engineering data of the same scale, the processing time of the present invention is only one-third of that of the traditional multi-software collaborative processing method.

[0122] (2) Accuracy:

[0123] Through precise geometric parameter calculation, strict direction determination and length calculation methods, as well as reasonable tolerance threshold setting, the accuracy of normal generation and data processing is ensured. In the process of attribute data fusion and node processing, scientific algorithms and filtering mechanisms are adopted to reduce the generation of errors. In practical applications, the accuracy of the generated cross-section data meets the requirements of engineering design. Through actual tests, its accuracy is equivalent to that of the multi-software collaborative processing method.

[0124] (3) Simple operation:

[0125] The user interaction interface built through the tkinter library is simple and intuitive to operate. The present invention has a total of three modules. In the normal generation module and the cross-section generation module, the user only needs to select the input file, set the relevant parameters and the output path, and then can start the data processing process with one key, without complex professional knowledge and operation skills; for the point cloud processing module, only one step of elevation pressing is required based on the existing ground points, reducing the usage threshold and improving the user experience.

[0126] (4) Standardized output:

[0127] The module directly outputs the cross-section in the Weidi format, ensuring that the cross-section data can be directly compatible with the user's required design software.

[0128] It should be understood that when each module of the extraction system for laser point cloud cross-section data provided in the above embodiments performs the extraction method for laser point cloud cross-section data, only the division of each functional module in the above description content is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0129] Each functional module in the above embodiments may be integrated into a processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the embodiments of the present application.

[0130] The above-described embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A system for extracting longitudinal and transverse section data of laser point cloud, characterized in that: include: A normal line production module, used for parsing the centerline data, generating a first normal line file with coordinate information and a second normal line file with attribute information; a point cloud processing module, configured to construct a surface model, project the first normal file onto the surface model to calculate three-dimensional coordinate information of the first normal file, and output a third normal file having the three-dimensional coordinate information; A section production module is used to generate one or more of cross-section data and longitudinal section data in a preset format according to the second normal file and the third normal file.

2. A method for extracting longitudinal and transverse section data of laser point cloud, characterized in that: include: Parsing the midline data to generate a first normal file with coordinate information and a second normal file with attribute information; constructing a surface model, projecting the first normal file onto the surface model to calculate three-dimensional coordinate information of the first normal file, and outputting a third normal file having the three-dimensional coordinate information; One or more of cross-section data and longitudinal section data in a preset format are generated according to the second normal file and the third normal file.

3. The method according to claim 2, characterized in that The method of analyzing the midline data includes: Extract geometric parameters and calculate line segment length; Connecting and integrating discrete line segments in the centerline data; Determine the direction of each line segment in the centerline data; Calculate the total length, offset and normal projection point of continuous line segments; The direction of the normal line is determined and the coordinates of the endpoints of the normal line are calculated.

4. The method according to claim 3, characterized in that The step of extracting geometric parameters and calculating line segment length includes: For a line segment entity or a polyline entity, the endpoint coordinates are extracted and the line segment length or the polyline length is calculated; for an arc entity or a circle entity, the center, radius, start and end angles are extracted, and the arc length and discrete point coordinates are calculated to extract geometric parameters and calculate the line segment length; The connecting and integrating discrete line segments in the centerline data includes: Construct a set of all line segment endpoints And set the tolerance threshold ε, if the two line endpoints meet If it is determined to be a continuous line segment, the adjacent line segments with common endpoints are connected; for a discontinuous path, the end point of the current path is connected with the starting point of the candidate line segment to connect and integrate the discrete line segments in the centerline data; Determining the direction of each line segment in the centerline data includes: Receive the starting point direction parameter, and determine the initial endpoint P of the continuous line segment according to the starting point direction parameter start and the termination endpoint P end The coordinate relationship of each line segment is determined according to the coordinate relationship, and the direction of the midline is normalized to be from the starting point to the end point to determine the direction of each line segment in the midline data; The calculation of the total length of the continuous line segments, the offset and the normal projection point includes: Calculating the Euclidean distance between consecutive points, and determining the total length of the consecutive line segments based on the Euclidean distance; Read the pile number from the engineering file and generate a pile number offset sequence, and determine a complete pile number set S based on the pile number offset sequence; For each pile number s i ∈S, calculate the cumulative length of the path s acc =s i -s0; then traverse the geometric path sub-segment {(p k ,p k+1 ,L k )}, locate the target point by segmented accumulation, when s prev ≤s acc ≤s prev +L k Calculate the local proportion in the case of Among them, s prev is the cumulative length of the preceding sequence, p k is the starting point of the sub-segment, P k+1 is the end point of the sub-segment, L k is the Euclidean distance of the sub-segment; the interpolation coordinates of the target point are calculated as (x, y) = (x k +λ(x k+1 -x k ), y k +λ(y k+1 -y k )) to calculate the tangent vector of the current sub-segment and determining the normal projection point after normalizing the tangent vector; The determining the direction of the normal line and calculating the endpoint coordinates of the normal line comprises: The left normal and the right normal are determined according to the normalized tangent vector, and the endpoint coordinates of the normal are determined according to the preset left and right buffer distances.

5. The method according to claim 2, characterized in that: The construction of the surface model comprises: A Delaunay triangulation network is constructed based on the input laser point cloud ground point data, and the surface model is constructed based on the Delaunay triangulation network.

6. The method according to claim 2, characterized in that The step of generating one or more of the cross-section data and the longitudinal section data in a preset format according to the second normal file and the third normal file comprises: Extracting three-dimensional polylines and three-dimensional coordinates of nodes on the polylines from the second normal file and the third normal file; Merging the attribute information of the input data with the extracted three-dimensional coordinates of the node so that the three-dimensional coordinates of the node are accompanied by corresponding attribute information; For nodes with the same normal, they are arranged in ascending order according to their distance from the midline projection; Calculate the plane distance and height difference between adjacent nodes; A filtering threshold is set, and the nodes are thinned out according to the filtering threshold, the plane distance, and the height difference.

7. The method according to claim 6, characterized in that The step of fusing the attribute information of the input data with the extracted three-dimensional coordinates of the node so that the three-dimensional coordinates of the node are accompanied by corresponding attribute information includes: Based on the matching relationship of the spatial positions, the plane coordinates of the node are intersected with the geographic space information in the second normal file to obtain the three-dimensional coordinates of the node with corresponding attribute information; The setting of the filtering threshold and performing thinning processing on the nodes according to the filtering threshold, the plane distance and the height difference include: Traverse the node sequence, when the plane distance D between adjacent nodes i When the distance is less than the minimum threshold ε, the two nodes are marked as coincident points and merged; when the height difference threshold ΔH i If the height difference is greater than the first height difference threshold, a manual review prompt message is sent; if the height difference threshold ΔH i When the height difference is less than the second high difference threshold, the node sequence less than the high difference threshold is deleted.

8. The method according to claim 6, characterized in that The calculating of the plane distance and height difference of adjacent nodes includes: For adjacent nodes p i =(x i ,y i , z i ) and p i+1 =(x i+1 ,y i+1 , z i+1 ), the plane distance is calculated according to the Euclidean distance. Calculate the vertical difference ΔH of adjacent nodes based on 3D coordinates i =Z i+1 -Z i .

9. The method according to claim 2, characterized in that: The file header of the preset format longitudinal section data is marked with HINTCAD5.83_DMX_SHUJU; each line records the pile number and its benchmark elevation.

10. The method according to claim 2, characterized in that The file header of the preset format cross-section data is identified by HINTCAD5.83_HDM_SHUJU; grouped by pile number, each pile number data block contains: pile number value, left / right side section points, plane distance change and height difference arranged alternately.

Images