Power line conductor point cloud data extraction method and system based on point cloud density

Abstract

A method and system for extracting conductor data from power distribution lines based on point cloud density is disclosed. The method includes: acquiring laser point cloud data of the power distribution line; preprocessing the acquired laser point cloud data to obtain segmented laser point cloud data of the power distribution line; clustering the segmented laser point cloud data of the power distribution line and using the noise point data obtained from the clustering as preliminary laser point cloud data of the power distribution conductor; projecting and filtering the preliminary laser point cloud data of the power distribution conductor to obtain laser point cloud data on both sides of the power distribution conductor; projecting the laser point cloud data on both sides of the power distribution conductor and clustering it based on an improved clustering algorithm to extract the laser point cloud data of the power distribution conductor; and obtaining the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor. This invention employs an improved clustering algorithm, starting from a global perspective and density features, to directly filter and finely extract the conductor point cloud data. The steps are simple and have a high fault tolerance.

Description

Technical Field

[0001] This invention belongs to the technical field of power distribution line conductor extraction, specifically relating to a method and system for power distribution line conductor extraction based on point cloud density. Background Technology

[0002] Extracting conductor coordinates from laser point cloud data of power distribution lines helps with UAV trajectory planning and provides digital empowerment for line operation and maintenance. The spatial characteristic of conductors is straight lines. Traditional straight-line detection algorithms, such as the Hough algorithm, LSD algorithm, and CannyLines algorithm, are only suitable for straight-line detection in two-dimensional images and are not applicable in three-dimensional space. While mature technologies exist for extracting power transmission conductors from laser point clouds of power transmission lines, the lower poles of distribution lines cause the point clouds of power lines and trees to blend together, whereas power transmission conductors are taller and have more distinct spatial features, making them completely different from distribution conductors. Therefore, methods used to extract power transmission conductors from laser point cloud data are no longer suitable for extracting distribution conductors.

[0003] In the prior art, patent document 1 proposes a method and apparatus for denoising 3D laser point cloud data in a railway scene. It mentions acquiring point cloud data along the railway line using a laser point cloud acquisition device, dividing the point cloud data into multiple point cloud data segments; acquiring multiple coordinate values ​​obtained by projecting each point cloud data segment onto a two-dimensional plane; dividing the two-dimensional plane into multiple grids, determining multiple grid index points corresponding to the multiple coordinate values; clustering the multiple grid index points using each grid index point vector as a cluster center to determine multiple initial categories of the grid index points; merging every two initial categories of the grid index points according to a hierarchical clustering algorithm to obtain multiple final categories of the grid index points and the number of grid index points contained in each final category; identifying noise points among the multiple grid index points, and identifying and removing the noise data corresponding to the noise points in the point cloud data segments, thus achieving fast and efficient denoising.

[0004] In the power sector, laser point cloud technology for transmission lines is relatively mature, but research on point cloud processing for distribution lines is limited. Existing technologies provide coarse positioning of towers within the distribution line grid, then detect conductor point clouds based on tower locations. However, the actual diameter of distribution towers is only 10-20 cm, making it very easy to miss tower point clouds during laser point cloud acquisition. For example, the lower half of the tower is often not captured. Therefore, only relatively complete towers can be located and acquired within the grid. Furthermore, there are instances where towers and trees are mixed together; existing technologies do not address how to distinguish between tower and tree point clouds in complex environments, leading to inaccurate extraction of tower conductors.

[0005] Due to the significant differences in transmission and distribution line corridors, most point cloud extraction techniques applied to transmission lines cannot be applied to distribution lines for the following reasons: 1. Transmission lines are generally over 30 meters high. After removing noise points, the lightning protection wire can be fitted based on the highest point to further determine the position of the conductor below. However, many conductors in power distribution channels are located inside trees, and the highest point is not necessarily the position of the conductor, so this method cannot be applied.

[0006] 2. Transmission lines have no debris around the conductors, so straight line fitting can be used to extract them. However, distribution lines have very complex channels, and there may be various trees around the conductors, making straight line fitting impossible.

[0007] 3. Transmission towers are much taller than the surrounding trees and have a very obvious vertical feature, so PCA principal component analysis can be used to determine the location of the towers. However, distribution line towers are mixed with trees and are of similar height, so PCA principal component analysis cannot distinguish between the towers and trees.

[0008] 4. The vertical symmetry of power transmission and distribution towers is very obvious, but power transmission towers are very tall, much taller than the surrounding trees, and can be directly distinguished from point clouds, while distribution towers cannot be directly distinguished from the surrounding trees.

[0009] 5. The DBSCAN algorithm is based on density clustering. Compared with conductors, trees, buildings, etc., the point cloud of transmission towers is more concentrated and the density is high enough. By adjusting the parameters, the point cloud of the tower can be completely distinguished. However, the volume of distribution towers is smaller, generally only tens of centimeters in diameter, and the point cloud is more scattered, and may even be classified as noise points. Summary of the Invention

[0010] To address the shortcomings of existing technologies, this invention provides a method and system for extracting power line conductors based on point cloud density, which can solve the technical problem of inaccurate conductor point cloud data extraction in existing technologies.

[0011] This invention proposes a method for extracting conductors of power distribution lines based on point cloud density, comprising: The laser point cloud data of the power distribution line is acquired, and the acquired laser point cloud data of the power distribution line is preprocessed to obtain segmented laser point cloud data of the power distribution line. Cluster the segmented laser point cloud data of the power distribution line, and use the noise point data obtained from the clustering as the preliminary laser point cloud data of the power distribution conductor; The preliminary laser point cloud data of the power distribution conductor is projected and filtered to obtain the laser point cloud data on both sides of the power distribution conductor. The laser point cloud data on both sides of the power distribution conductor is projected and clustered based on an improved clustering algorithm to extract the laser point cloud data of the power distribution conductor. The coordinates of the power distribution conductor are then obtained based on the laser point cloud data of the power distribution conductor.

[0012] Preferably, the acquisition of laser point cloud data of the power distribution line involves preprocessing the acquired laser point cloud data of the power distribution line. The preprocessing includes segmenting the power distribution line to obtain segmented laser point cloud data of the power distribution line, specifically including: Collect laser point cloud data of power distribution lines, including the three-dimensional coordinates and reflection intensity of the power distribution lines; Based on the three-dimensional coordinate data of the power distribution line, the power distribution line is divided into segments according to the line direction to obtain the data of each segment of the power distribution line. Calculate the quartiles of the z-coordinate of each segment of the power distribution line data, and filter out laser point cloud data with z values ​​greater than the 75th quartile. Based on the reflection intensity of the laser point cloud data, laser point cloud data with a reflection intensity greater than 0 are selected from those with a z-value greater than the 75th quartile and used as the segmented laser point cloud data for power distribution lines.

[0013] Preferably, the step of clustering the segmented laser point cloud data of the power distribution line and using the clustered noise point data as the preliminary laser point cloud data of the power distribution conductor specifically includes: Obtain the three-dimensional coordinates of each segment of the laser point cloud data in the segmented laser point cloud data of the power distribution line; Clustering parameters were set based on the laser point cloud acquisition density and the spatial characteristics of power line conductors, towers, and ground point clouds. The DBSCAN algorithm was used to perform three-dimensional clustering of the three-dimensional coordinates of the laser point cloud data. The obtained clustering results include point cloud data of noise points with a label of -1. The point cloud data of noise points is used as the preliminary laser point cloud data of the power distribution conductor.

[0014] Preferably, the setting of clustering parameters specifically includes: setting the first neighborhood radius eps to 0.5 and the first core point threshold min_samples to 200.

[0015] Preferably, the projection and filtering of the preliminary laser point cloud data of the power distribution conductor to obtain laser point cloud data on both sides of the power distribution conductor specifically includes: The preliminary laser point cloud data of each segment of the power distribution conductor is projected onto the xoy plane; The linear equations of the traverse directions of each segment are obtained by fitting the projection results, as well as the angle θ between the linear equations and the x-axis. Based on the linear equation of the direction of each segment of the conductor, the preliminary laser point cloud data of each segment of the conductor is rotated clockwise by an angle θ to obtain preliminary laser point cloud data parallel to the x-axis. Based on the preliminary laser point cloud data parallel to the x-axis, the minimum value y_min and the maximum value y_max of the y-coordinate of this segment of laser point cloud data are obtained, and the center line position y_mean = (y_max + y_min) / 2 is calculated. Using the centerline as a reference, the laser point cloud data within 2 meters on both sides of the centerline are selected, and noise points on the side of the line channel are filtered out to obtain the laser point cloud data on both sides of the power distribution conductor.

[0016] Preferably, the step of projecting the laser point cloud data on both sides of the power distribution conductor and clustering it based on an improved clustering algorithm to extract the laser point cloud data of the power distribution conductor, and obtaining the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor, specifically includes: Project the laser point cloud data on both sides of the power distribution conductor onto the yoz plane; Based on the projection results of the yoz plane, the parameters of the improved DBSCAN algorithm are calculated, and the improved DBSCAN algorithm is used to perform three-dimensional clustering on the projection results to obtain the clustering results. Based on the clustering results, a single conductor converges into a single point on the yoz plane, with scattered noise points around it. The noise points are labeled -1. The point cloud data labeled -1 are removed to obtain the laser point cloud data of the power distribution conductor. The obtained three-dimensional laser point cloud data of the power distribution conductor is rotated counterclockwise by an angle θ in the coordinate system to obtain the three-dimensional coordinates of the power distribution conductor in the original coordinate system.

[0017] Preferably, the calculation of the improved DBSCAN algorithm parameters specifically includes: The second neighborhood radius and the second core point threshold of the DBSCAN algorithm are calculated based on the projection results. The calculation formula is as follows;

[0018] In the formula, This represents the radius of the DBSCAN neighborhood on the yoz projection plane. Indicates the basic proportionality coefficient. This represents the span of the projected point cloud along the y-axis. This represents the span of the projected point cloud along the z-axis. Indicates the density influence coefficient. This represents the point cloud density of the current projection area. Indicates the maximum expected density; Second core point threshold The calculation formula is as follows:

[0019] In the formula, This represents the minimum number of samples for the DBSCAN core points on the yoz projection plane. This represents the proportion of the total number of points to the minimum base sample size. This indicates the total number of point clouds in the current projection area. Indicates average density, This represents the power of the density exponent, controlling the degree to which density affects the minimum sample size.

[0020] The present invention also proposes a power distribution line conductor extraction system based on point cloud density, for implementing the power distribution line conductor extraction method based on point cloud density, including: a data acquisition module, a preprocessing module, a clustering module, and a projection clustering module; The data acquisition module is used to acquire laser point cloud data of power distribution lines; The preprocessing module is used to preprocess the acquired laser point cloud data of the power distribution line to obtain segmented laser point cloud data of the power distribution line. The clustering module is used to cluster the obtained segmented laser point cloud data of the power distribution line to obtain the preliminary laser point cloud data of the power distribution conductor; The projection clustering module is used to project and filter the preliminary laser point cloud data of the power distribution conductor to obtain the laser point cloud data on both sides of the power distribution conductor; and to project the laser point cloud data on both sides of the power distribution conductor and cluster it based on the improved clustering algorithm to extract the laser point cloud data of the power distribution conductor, and to obtain the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor.

[0021] The present invention also proposes a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the point cloud density-based power line conductor extraction method.

[0022] The present invention also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method for extracting power line conductors based on point cloud density.

[0023] The beneficial effects of this invention are as follows: Starting from the density characteristics of power distribution conductors, this invention initially extracts conductor coordinates based on these density characteristics, and then further refines the conductor coordinates based on the spatial characteristics of the power distribution conductors. This invention employs the DBSCAN clustering algorithm, using the noise points obtained from clustering as the initial power distribution conductor point cloud. Then, through projection and an improved clustering algorithm, noise points in the initial power distribution conductor point cloud are further removed, ultimately extracting the three-dimensional coordinates of the power distribution conductors. This enables accurate extraction of power distribution line conductors in complex surrounding environments. In traditional DBSCAN clustering applications, the neighborhood radius and minimum sample number parameters are usually set to fixed values ​​based on experience or multiple trials. However, this invention dynamically combines the clustering parameters with the spatial geometric features (span) and statistical features (density) of the current point cloud data and calculates the neighborhood radius and minimum sample number parameters. This allows for adaptation to the complex conditions of sparse, unevenly distributed, and variable scene conditions of conductor point clouds in power distribution lines, significantly improving the accuracy, robustness, and generalization ability of conductor point cloud extraction to different line environments. This invention starts from a global perspective and density characteristics, directly selecting and finely extracting conductor point clouds, with simple steps and high fault tolerance.

[0024] This invention filters out noise points along the line channel by selecting laser point cloud data within 2 meters on both sides of the center line, and performs noise reduction through two-dimensional planar projection. By removing point clouds of trees, buildings, etc. that are easy to cluster in the power distribution line point cloud, the conductor point cloud that belongs to noise points is left. Further projection clustering and fine extraction are then performed to successfully extract the power distribution conductor point cloud from the power distribution line point cloud. This method is more suitable for power distribution line point cloud processing in power scenarios. Attached Figure Description

[0025] Figure 1 This is a flowchart of the power distribution line conductor extraction method based on point cloud density in this invention; Figure 2 These are the preliminary screening results from DBSCAN clustering; Figure 3 This is the result of the yoz plane projection; Figure 4 It is a 3D visualization result of power distribution conductors; Figure 5 This is a structural diagram of the power distribution line conductor extraction system based on point cloud density in this invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, other embodiments obtained by those skilled in the art without creative effort are all within the protection scope of this invention.

[0027] Example 1 like Figure 1 As shown, this invention proposes a method for extracting power line conductors based on point cloud density, which includes the following steps: Step 1: Obtain laser point cloud data of the power distribution line. Preprocess the obtained laser point cloud data of the power distribution line. The preprocessing includes segmenting the power distribution line to obtain segmented laser point cloud data of the power distribution line. Specifically, step 1 includes: Step 1.1: Use the laspy library in Python to read the laser point cloud data of the power distribution line and obtain the x, y, z three-dimensional coordinate data and intensity reflection intensity data of each point cloud data. Step 1.2: Based on the three-dimensional coordinates of the power distribution line obtained in Step 1.1, the line is divided along its direction. Each segment is approximately 100 meters long and is a straight line without any corner towers. This yields the laser point cloud data of the divided power distribution line. Step 1.3: Based on the segmentation results of Step 1.2, calculate the quartiles of the z-coordinate of the laser point cloud data in each segment, and filter out the laser point cloud data with z values ​​greater than the 75th quartile. The quartiles of the z-coordinate include: minimum value, value at 25%, value at 50%, value at 75%, and maximum value. Step 1.4: Based on the screening results of Step 1.3, further filter out laser point cloud data with intensity reflection greater than 0 to obtain segmented laser point cloud data of power distribution lines.

[0028] Step 2: Cluster the segmented laser point cloud data of the power distribution line obtained in Step 1 to obtain the preliminary laser point cloud data of the power distribution conductor. Specifically, step 2 includes: Step 2.1: Based on the segmented laser point cloud data of the power distribution line obtained in Step 1, obtain the x, y, and z three-dimensional coordinates of each segment of the laser point cloud data; Step 2.2: The DBSCAN algorithm is used to perform 3D clustering on the 3D coordinates obtained in Step 2.1. Based on the laser point cloud acquisition density and the spatial characteristics of power line conductors, towers, and ground point clouds, the following parameters are set: the first neighborhood radius eps is 0.5, the first core point threshold min_samples is 200. The neighborhood radius is used as a distance threshold to determine whether two sample points belong to the same cluster, and the core point threshold represents the minimum number of neighbors required for a point to become a core point. In this invention, it is preferred that the distance between each element in the cluster is less than 0.5 meters and the number of elements in the cluster is greater than 200. like Figure 2 As shown, Figure 2 Preliminary screening results for DBSCAN clustering.

[0029] Step 2.3: Based on the clustering results in Step 2.2, compared to the point clouds of the ground, trees, etc., the point cloud of the power distribution line conductors is very sparse and is classified as noise points in the clustering. The noise point label is -1. Therefore, the point cloud data with the label -1 is extracted to obtain the preliminary laser point cloud data of the power distribution line conductors.

[0030] Step 3: Project and filter the preliminary laser point cloud data of the power distribution conductor obtained in Step 2. After filtering the noise data on both sides of the power distribution conductor, the laser point cloud data on both sides of the power distribution conductor is obtained. Clustering yields the three-dimensional coordinates of the power distribution conductors.

[0031] Specifically, step 3 includes: Step 3.1: Project the preliminary laser point cloud data of each conductor segment obtained in Step 2 onto the xoy plane; Step 3.2: Based on the projection results obtained in Step 3.1, fit the linear equation of the traverse direction, with an angle of θ between it and the x-axis; Step 3.3: Based on the linear equation of the conductor direction obtained in Step 3.2, rotate the preliminary laser point cloud data of each conductor segment obtained in Step 3 clockwise by an angle θ to obtain preliminary laser point cloud data parallel to the x-axis. Step 3.4: Based on the preliminary laser point cloud data parallel to the x-axis obtained in Step 3.2, calculate the minimum value y_min, the maximum value y_max, and the center line position y_mean of the laser point cloud data segment, i.e. (y_max + y_min) / 2. Using the center line as the reference, shift 2 meters upwards and downwards respectively, and select the laser point cloud data within 2 meters on both sides of the center line as the laser point cloud data on both sides of the power distribution conductor, and filter out the laser point cloud data outside 2 meters on both sides of the center line. Step 4: Project the laser point cloud data on both sides of the power distribution conductor, calculate the clustering algorithm parameters based on the projection results, perform three-dimensional clustering on the projection results based on the improved parameter clustering algorithm, extract the laser point cloud data of the power distribution conductor, and obtain the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor.

[0032] Because the point cloud density of power distribution line conductors is sparse, the point cloud data of power distribution line conductors cannot be directly clustered. This invention first obtains noise points through density-based DBSCAN algorithm clustering, and then filters the noise points. Step 4 specifically includes: Step 4.1: Project the laser point cloud data within 2 meters on both sides of the center line obtained in Step 3.4 onto the yoz plane; like Figure 3 As shown, Figure 3 This is a schematic diagram of the projection result of the yoz plane.

[0033] Step 4.2: Calculate the improved DBSCAN algorithm parameters based on the projection results obtained in Step 4.1, and use the DBSCAN algorithm to perform three-dimensional clustering on the projection results; Specifically, the calculation of the improved DBSCAN algorithm parameters includes: The second neighborhood radius and the second core point threshold of the DBSCAN algorithm are calculated based on the projection results. The calculation method is as follows: Second neighborhood radius The calculation formula is as follows;

[0034] In the formula, This represents the radius of the DBSCAN neighborhood on the yoz projection plane. Indicates the basic proportionality coefficient. This represents the span of the projected point cloud along the y-axis. This represents the span of the projected point cloud along the z-axis. Indicates the density influence coefficient. This represents the point cloud density of the current projection area. Indicates the maximum expected density; Second core point threshold The calculation formula is as follows:

[0035] In the formula, This represents the minimum number of samples for the DBSCAN core points on the yoz projection plane. This represents the proportion of the total number of points to the minimum base sample size. This indicates the total number of point clouds in the current projection area. Indicates average density, This represents the power of the density exponent, controlling the degree to which density affects the minimum sample size.

[0036] Step 4.3: Based on the clustering results obtained in Step 4.2, a single conductor can basically converge into a single point on the yoz plane, with scattered noise points around it. The label of the noise points is -1. Therefore, the point cloud data with the label -1 is removed to obtain the laser point cloud data of the power distribution conductor. Step 4.4: Based on the laser point cloud data of the power distribution conductor obtained in Step 4.3, rotate counterclockwise by an angle θ to obtain the three-dimensional coordinates of the power distribution conductor in the original coordinate system.

[0037] like Figure 4 The image shown is a schematic diagram of the obtained three-dimensional visualization results of the power distribution conductors.

[0038] Example 2 like Figure 5 As shown, this invention also proposes a power distribution line conductor extraction system based on point cloud density, used to implement the power distribution line conductor extraction method based on point cloud density proposed in Example 1. The system includes: Data acquisition module, preprocessing module, clustering module, and projection clustering module; The data acquisition module is used to acquire laser point cloud data of power distribution lines; The preprocessing module is used to preprocess the acquired laser point cloud data of the power distribution line to obtain segmented laser point cloud data of the power distribution line. The clustering module is used to cluster the obtained segmented laser point cloud data of the power distribution line to obtain the preliminary laser point cloud data of the power distribution conductor; The projection clustering module is used to project and filter the preliminary laser point cloud data of the power distribution conductor to obtain the laser point cloud data on both sides of the power distribution conductor; and to project the laser point cloud data on both sides of the power distribution conductor and cluster it based on the improved clustering algorithm to extract the laser point cloud data of the power distribution conductor, and to obtain the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor.

[0039] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0040] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0041] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0042] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for extracting point cloud data of power distribution line conductors based on point cloud density, characterized in that, include: The laser point cloud data of the power distribution line is acquired, and the acquired laser point cloud data of the power distribution line is preprocessed to obtain segmented laser point cloud data of the power distribution line. Cluster the segmented laser point cloud data of the power distribution line, and use the noise point data obtained from the clustering as the preliminary laser point cloud data of the power distribution conductor; The preliminary laser point cloud data of the power distribution conductor is projected and filtered to obtain the laser point cloud data on both sides of the power distribution conductor. The laser point cloud data on both sides of the power distribution conductor is projected and clustered based on an improved clustering algorithm to extract the laser point cloud data of the power distribution conductor. The coordinates of the power distribution conductor are then obtained based on the laser point cloud data of the power distribution conductor.

2. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 1, characterized in that: The process of acquiring laser point cloud data of power distribution lines involves preprocessing the acquired laser point cloud data of power distribution lines. This preprocessing includes segmenting the power distribution lines to obtain segmented laser point cloud data of the power distribution lines. Specifically, this includes: Collect laser point cloud data of power distribution lines, including the three-dimensional coordinates and reflection intensity of the power distribution lines; Based on the three-dimensional coordinate data of the power distribution line, the power distribution line is divided into segments according to the line direction to obtain the data of each segment of the power distribution line. Calculate the quartiles of the z-coordinate of each segment of the power distribution line data, and filter out laser point cloud data with z values ​​greater than the 75th quartile. Based on the reflection intensity of the laser point cloud data, laser point cloud data with a reflection intensity greater than 0 are selected from those with a z-value greater than the 75th quartile and used as the segmented laser point cloud data for power distribution lines.

3. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 1, characterized in that: The process of clustering segmented laser point cloud data of power distribution lines and using the resulting noise point data as preliminary laser point cloud data for power distribution conductors specifically includes: Obtain the three-dimensional coordinates of each segment of the laser point cloud data in the segmented laser point cloud data of the power distribution line; Clustering parameters were set based on the laser point cloud acquisition density and the spatial characteristics of power line conductors, towers, and ground point clouds. The DBSCAN algorithm was used to perform three-dimensional clustering of the three-dimensional coordinates of the laser point cloud data. The obtained clustering results include point cloud data of noise points with a label of -1. The point cloud data of noise points is used as the preliminary laser point cloud data of the power distribution conductor.

4. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 3, characterized in that: The specific clustering parameter settings include: setting the first neighborhood radius eps to 0.5 and the first core point threshold min_samples to 200.

5. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 1, characterized in that: The process of projecting and filtering the preliminary laser point cloud data of the power distribution conductor to obtain laser point cloud data on both sides of the power distribution conductor specifically includes: The preliminary laser point cloud data of each segment of the power distribution conductor is projected onto the xoy plane; The linear equations of the traverse directions of each segment are obtained by fitting the projection results, as well as the angle θ between the linear equations and the x-axis. Based on the linear equation of the direction of each segment of the conductor, the preliminary laser point cloud data of each segment of the conductor is rotated clockwise by an angle θ to obtain preliminary laser point cloud data parallel to the x-axis. Based on the preliminary laser point cloud data parallel to the x-axis, the minimum value y_min and the maximum value y_max of the y-coordinate of this segment of laser point cloud data are obtained, and the center line position y_mean = (y_max + y_min) / 2 is calculated. Using the centerline as a reference, the laser point cloud data within 2 meters on both sides of the centerline are selected, and noise points on the side of the line channel are filtered out to obtain the laser point cloud data on both sides of the power distribution conductor.

6. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 1, characterized in that: The process of projecting the laser point cloud data on both sides of the power distribution conductor and clustering it based on an improved clustering algorithm to extract the laser point cloud data of the power distribution conductor, and then obtaining the coordinates of the power distribution conductor based on the laser point cloud data, specifically includes: Project the laser point cloud data on both sides of the power distribution conductor onto the yoz plane; Based on the projection results of the yoz plane, the parameters of the improved DBSCAN algorithm are calculated, and the improved DBSCAN algorithm is used to perform three-dimensional clustering on the projection results to obtain the clustering results. Based on the clustering results, a single conductor converges into a single point on the yoz plane, with scattered noise points around it. The noise points are labeled -1. The point cloud data labeled -1 are removed to obtain the laser point cloud data of the power distribution conductor. The obtained three-dimensional laser point cloud data of the power distribution conductor is rotated counterclockwise by an angle θ in the coordinate system to obtain the three-dimensional coordinates of the power distribution conductor in the original coordinate system.

7. The method for extracting point cloud data of power distribution line conductors based on point cloud density according to claim 6, characterized in that: The calculation of the improved DBSCAN algorithm parameters specifically includes: The second neighborhood radius and the second core point threshold of the DBSCAN algorithm are calculated based on the projection results. The calculation formula is as follows; In the formula, This represents the radius of the DBSCAN neighborhood on the yoz projection plane. Indicates the basic proportionality coefficient. This represents the span of the projected point cloud along the y-axis. This represents the span of the projected point cloud along the z-axis. Indicates the density influence coefficient. This represents the point cloud density of the current projection area. Indicates the maximum expected density; Second core point threshold The calculation formula is as follows: In the formula, This represents the minimum number of samples for the DBSCAN core points on the yoz projection plane. This represents the proportion of the total number of points to the minimum base sample size. This indicates the total number of point clouds in the current projection area. Indicates average density, This represents the power of the density exponent, controlling the degree to which density affects the minimum sample size.

8. A point cloud data extraction system for power distribution line conductors based on point cloud density, used to implement the point cloud data extraction method for power distribution line conductors based on point cloud density as described in any one of claims 1-7, characterized in that, include: Data acquisition module, preprocessing module, clustering module, and projection clustering module; The data acquisition module is used to acquire laser point cloud data of power distribution lines; The preprocessing module is used to preprocess the acquired laser point cloud data of the power distribution line to obtain segmented laser point cloud data of the power distribution line. The clustering module is used to cluster the obtained segmented laser point cloud data of the power distribution line to obtain the preliminary laser point cloud data of the power distribution conductor; The projection clustering module is used to project and filter the preliminary laser point cloud data of the power distribution conductor to obtain the laser point cloud data on both sides of the power distribution conductor; and to project the laser point cloud data on both sides of the power distribution conductor and cluster it based on the improved clustering algorithm to extract the laser point cloud data of the power distribution conductor, and to obtain the coordinates of the power distribution conductor based on the laser point cloud data of the power distribution conductor.

9. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-7.

Images