A method and system for adaptive segmentation of a line channel of an unmanned airborne laser radar

Abstract

This invention provides an adaptive segmentation method and system for unmanned aerial vehicle (UAV)-borne lidar (UAV) lane channels. The adaptive segmentation method includes: real-time computation of lidar point cloud data to extract multiple power line point clouds; fitting the power line line equations corresponding to each power line point cloud; calculating multiple power line lines using the power line line equations; clustering parallel power line lines into the same power line group using a parallel line clustering algorithm; dividing the same power line group into multiple lane sets; calculating the minimum line spacing of the power line lines; segmenting the remaining power line lines into lane sets based on the minimum line spacing and the average line spacing of the lane sets; and aggregating all power line lines from the same lane set into the same lane channel when all power line lines have been segmented. This invention solves the problem of existing technologies struggling to automatically identify all lane channels and to automatically follow user-intended lane paths.

Description

Technical Field

[0001] This invention relates to the field of lidar technology, and in particular to an adaptive segmentation method and system for line channels of UAV-borne lidar. Background Technology

[0002] LiDAR is widely used in the field of automated power line inspection. LiDAR acquires point cloud data of power lines to achieve inspection objectives. Utilizing drones equipped with LiDAR for automated power line inspection can effectively reduce workload, increase the level of automation, and facilitate stable aerial inspection of power lines.

[0003] During UAV line-following flight, it's inevitable to encounter situations where multiple line channels exist within a predetermined distance below or in front of the UAV. To select the appropriate line channel for line-following, all line channels must be correctly identified beforehand to enable automatic line-following flight according to the user's intent. Especially in scenarios with numerous parallel lines, or even simultaneous parallel main and distribution networks, significant uncertainty arises during automatic flight. This can ultimately lead to tracking the wrong channel or deviating from the required flight path, impacting data acquisition. Summary of the Invention

[0004] This invention provides an adaptive segmentation scheme for the line channel of UAV-borne lidar, which aims to solve the problem that in the prior art, there are multiple line channels, and the UAV has great uncertainty during automatic flight, which leads to tracking errors or deviation from the channel area and affects the UAV's data collection operation.

[0005] To address the aforementioned problems, according to a first aspect of the present invention, an adaptive segmentation method for the line channel of an unmanned aerial vehicle (UAV)-borne lidar is proposed, comprising:

[0006] During the drone's flight, the point cloud data collected by the lidar is processed in real time to extract multiple power line point clouds;

[0007] According to the predetermined line fitting algorithm, the equations of the power line lines corresponding to each power line point cloud in multiple power line point clouds are fitted respectively.

[0008] The electric field line equation is used to select any electric field line as a reference line. The reference line is used to correct the direction of other electric field lines. Parallel electric field lines are clustered into the same electric field group according to the parallel line clustering algorithm.

[0009] Based on the spacing between the straight power lines, the same power line is divided into multiple line sets.

[0010] For each of the multiple line sets, calculate the minimum line spacing between the remaining power line straight lines in the same power line group and the line set. Based on the relationship between the minimum line spacing and the average line spacing of the channel in the line set, divide the remaining power line straight lines into the line set.

[0011] When all power line lines in the same power line group have been separated, all power line lines in the same line set are aggregated into the same line channel.

[0012] Preferably, in the above-mentioned adaptive segmentation method for power line channels, the step of extracting multiple power line point clouds by real-time calculation of the point cloud data collected by lidar includes:

[0013] Use lidar to collect point cloud data of the target power line area;

[0014] The point cloud data is calculated based on the relationship between the lidar coordinate system and the world coordinate system to obtain the point cloud data in the world coordinate system.

[0015] Based on a predetermined point cloud feature extraction algorithm, multiple power line point clouds are extracted from point cloud data in the world coordinate system.

[0016] Preferably, in the above-mentioned adaptive segmentation method for power line channels, the step of fitting the power line straight line equation corresponding to each power line point cloud in multiple power line point clouds according to a predetermined power line fitting algorithm includes:

[0017] The RANSAC algorithm is used to select the coordinates of any two laser points in the point cloud of each power line;

[0018] The fitted center line of the electric field point cloud is obtained by calculating the coordinates of any two laser points;

[0019] Based on the formula for calculating straight-line distance, the distances from other laser points in the electric field point cloud to the fitted center line are obtained respectively;

[0020] Count the number of other laser points whose distance from the fitted center line is less than or equal to the predetermined shortest distance value;

[0021] Recalculate the fitted center line of the electric field point cloud by selecting the coordinates of any two laser points. Continue until the number of other laser points whose distance from the fitted center line is less than or equal to the predetermined shortest distance value is the largest. Then, use the recalculated fitted center line as the electric field line equation corresponding to the electric field point cloud.

[0022] Preferably, in the above-mentioned adaptive segmentation method for power line corridors, the steps of selecting any power line straight line as a reference line using the power line straight line equation, using the reference line to correct the direction of the seven-day power line straight lines, and clustering mutually parallel power line straight lines into the same power line group according to the parallel line clustering algorithm include:

[0023] The electric field line straight line corresponding to the electric field line point cloud is calculated using the electric field line straight line equation.

[0024] Select any one of the power line lines from all the power line lines as the reference line for the power line group, and calculate the angle between the other power line lines and the reference line in all the power line lines respectively.

[0025] Determine whether the angle between other power lines and the reference line is less than or equal to a predetermined angle threshold;

[0026] If the included angle of the power lines is less than or equal to a predetermined included angle threshold, then other power lines are determined to be parallel to the reference line, and the other power lines are added to the power line group where the reference line is located.

[0027] If the included angle of the electric field lines is greater than a predetermined included angle threshold, then any electric field line with an included angle greater than the predetermined included angle threshold is selected as a reference line. The process of determining whether the included angles of other electric field lines with the reference line are less than or equal to the predetermined included angle threshold is repeated until all parallel electric field lines are clustered into the same electric field line group.

[0028] Preferably, the above-mentioned adaptive segmentation method for power line channels, after the step of clustering mutually parallel power lines into the same power line group according to the parallel line clustering algorithm, further includes: removing duplicate or split power lines in the same power line group based on the distance between the power lines in the same power line group; this step of removing duplicate or split power lines in the same power line group based on the distance between the power lines in the same power line group includes:

[0029] Select any straight line of the electric field in the same group of electric fields as the reference line of the container, and calculate the distance between all other straight lines of the electric field in the same group of electric fields and the reference line.

[0030] Determine whether the distance between each other power line and the reference line is less than or equal to the predetermined straight line distance;

[0031] If the spacing is less than or equal to the predetermined straight line distance, then other power lines are identified as power line lines that repeat or split from the reference line, and the other power line lines are added to the container corresponding to the reference line.

[0032] If the spacing is greater than the predetermined straight line distance, then select any one of the power line lines with a spacing greater than the predetermined straight line distance as the reference line of the container, and repeat the step of determining whether the spacing between other power line lines and the reference line is less than or equal to the predetermined straight line distance, until all power line lines in the same power line group are added to the container of the corresponding reference line, wherein each reference line has a unique container.

[0033] Traverse all the power lines in all containers, select the highest power line from each container and merge them to obtain a group of identical power lines after deduplication.

[0034] Preferably, the above-mentioned adaptive segmentation method for power line channels further includes, after the step of clustering mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm:

[0035] Based on the spacing between the dividing lines of the power line straight line, determine whether there are multiple line channels in the same power line group;

[0036] If there are multiple line channels in the same power line group, then the step of dividing the same power line group into multiple line sets according to the line spacing of the power line straight line is performed;

[0037] If there are no multiple line channels within the same power line group, then all power line lines in the same power line group will be aggregated into the same line channel.

[0038] Preferably, in the above-mentioned adaptive segmentation method for power line channels, the step of calculating the minimum line spacing between the remaining power line lines in the same power line group and the power line group for each power line group among multiple power line sets, and segmenting the remaining power line lines into the power line group according to the relationship between the minimum line spacing and the average line spacing of the power line group, includes:

[0039] Select a straight power line from the power line group and add it to the currently segmented line set;

[0040] Calculate the minimum line spacing between each remaining power line straight line and the line set within the same power line group;

[0041] Determine whether the minimum line spacing between each remaining power line straight line and the line set in the same power line group is less than the line spacing of the segmented lines.

[0042] If the minimum line spacing is less than the segmented line spacing, then determine whether the minimum line spacing is less than the product of the average line spacing of the channel and the channel spacing factor of the line set.

[0043] If the minimum line spacing is less than the product of the average line spacing and the channel spacing factor in the line set, then the remaining power line straight lines are added to the line set.

[0044] If the minimum line spacing is greater than or equal to the product of the average line spacing and the channel spacing factor in the line set, then the remaining power line straight lines are retained to the same power line group.

[0045] Increment the spacing between the split lines and re-execute the step of determining whether the minimum line spacing between each remaining power line straight line and the line set in the same power line group is less than the split line spacing, until the split line spacing is greater than the preset maximum line spacing.

[0046] Once all power lines in the same power line group have been separated into straight lines, all power lines in the same line set will be aggregated into the same line channel.

[0047] Preferably, in the above-mentioned adaptive segmentation method for power line corridors, the step of calculating the minimum line spacing between each remaining power line straight line and the line set within the same power line group includes:

[0048] The straight lines of the power lines selected from the power line groups will be used as the initial straight lines of the line set;

[0049] Calculate the distance between each remaining power line line in the same power line group and the initial line of the line set, and use this distance as the minimum line spacing; or,

[0050] Calculate the distances between each remaining power line line in the same power line group and all lines in the line set;

[0051] The minimum distance between each remaining power line and all the lines in the line set is selected as the minimum line spacing.

[0052] When the spacing between the split lines is greater than the preset maximum spacing between the lines, determine whether there are any remaining straight power lines in the same power line group.

[0053] If there are remaining straight power lines in the same power line group, then all remaining straight power lines in the same power line group will be merged into an independent line corridor.

[0054] According to a second aspect of the present invention, the present invention also provides a line channel adaptive segmentation system for UAV-borne lidar, comprising:

[0055] The power line point cloud extraction module is used to process the point cloud data collected by the lidar in real time during the flight of the UAV and extract multiple power line point clouds.

[0056] The power line straight line equation fitting module is used to fit the power line straight line equation corresponding to each power line point cloud in multiple power line point clouds according to a predetermined line fitting algorithm.

[0057] The power line line clustering module is used to select any power line line as a reference line using the power line line equation, use the reference line to correct the direction of other power line lines, and cluster mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm.

[0058] The line set segmentation module is used to group the same power line into multiple line sets based on the line spacing of the power line straight lines;

[0059] The minimum line spacing calculation module is used to calculate the minimum line spacing between the remaining power line straight lines and the line set in the same power line group for each of multiple line sets.

[0060] The power line straight line segmentation module is used to segment the remaining power lines into the line set based on the relationship between the minimum line spacing and the average line spacing of the channel in the line set;

[0061] The power line straight line clustering module is used to aggregate all power line straight lines of the same line set into the same line channel after all power line straight lines in the same power line group have been segmented.

[0062] According to a third aspect of the present invention, the present invention also provides a line channel adaptive segmentation system for UAV-borne LiDAR, comprising: a memory, a processor, and a line channel adaptive segmentation program for UAV-borne LiDAR stored in the memory and running on the processor, wherein the line channel adaptive segmentation program implements the steps of the line channel adaptive segmentation method provided by any of the above technical solutions when the tracking channel adaptive segmentation program is executed by the processor.

[0063] In summary, the adaptive segmentation scheme for UAV-borne LiDAR lines provided in this application extracts multiple power line point clouds from the point cloud data collected by the LiDAR during UAV flight. Then, based on a predetermined line fitting algorithm, the equations of the corresponding power line lines for each power line point cloud are fitted. These equations are then used to calculate the power line lines, facilitating their placement into the corresponding line channels. After obtaining the power line lines for each power line point cloud, a parallel line clustering algorithm is used to cluster parallel power line lines into the same power line group, thus completing the initial segmentation. Since the same power line group may include multiple line channels, the power line group is divided into multiple line sets based on the line spacing of the power line lines. For each line set, the minimum line spacing between the remaining power line lines in the same power line group and the line set is calculated. Based on the relationship between this minimum line spacing and the average line spacing of the line set's channels, the remaining power line lines are segmented into the corresponding line sets. Therefore, all power line lines in the same line set can be aggregated into the same line channel. The above method enables adaptive segmentation of multiple line channels, effectively dividing multiple channels below the UAV and facilitating UAV flight along designated line channels during flight, thus achieving automated data collection of the channels. This adaptive line channel segmentation method can handle conventional and parallel lines in the main network and distribution network, as well as complex parallel lines adjacent to the main network and distribution network. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0065] Figure 1 This is a flowchart illustrating an adaptive segmentation method for the line channel of an unmanned aerial vehicle (UAV)-borne lidar, provided in an embodiment of the present invention.

[0066] Figure 2 yes Figure 1 The illustrated embodiment provides a flowchart of a method for extracting point clouds of multiple power lines;

[0067] Figure 3 yes Figure 1 The illustrated embodiment provides a flowchart of a method for fitting the equation of a straight electric field line.

[0068] Figure 4 yes Figure 1The illustrated embodiment provides a flowchart of a clustering method for straight power lines;

[0069] Figure 5 yes Figure 1 The illustrated embodiment provides a flowchart of a method for segmenting a line channel;

[0070] Figure 6 yes Figure 5 The illustrated embodiment provides a flowchart of a method for calculating minimum line spacing;

[0071] Figure 7 This is a flowchart illustrating a method for deduplicating straight power lines according to an embodiment of the present invention.

[0072] Figure 8 This is a flowchart illustrating a method for determining multiple line channels provided in an embodiment of the present invention;

[0073] Figure 9 This is a schematic diagram of the structure of a line channel adaptive segmentation system for an unmanned aerial vehicle (UAV)-borne lidar provided in an embodiment of the present invention;

[0074] Figure 10 This is a schematic diagram of the structure of an adaptive segmentation system for a UAV-borne lidar line channel provided in an embodiment of the present invention.

[0075] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0076] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0077] The main technical problem solved by the embodiments of the present invention is:

[0078] In existing technologies, during UAV line-following flight, situations inevitably arise where multiple line channels exist within a predetermined distance below or in front of the UAV. To select the appropriate line channel for line-following, all line channels must be correctly identified in advance to enable automatic line-following flight according to user intent. This is especially problematic in scenarios with numerous parallel lines, or even simultaneous parallel main and distribution networks, leading to significant uncertainty during automatic flight. Ultimately, this can result in tracking the wrong channel or deviating from the intended flight path, impacting data acquisition.

[0079] To address the aforementioned issues, the following embodiments of the present invention provide an adaptive segmentation scheme for power line channels using UAV-borne LiDAR. By utilizing UAV-borne LiDAR technology, laser point clouds are calculated in real-time during UAV flight, and the equations of straight power lines in the scene are extracted. By calculating whether these power line equations are parallel, power lines are clustered into the same power line group, thus achieving initial segmentation of the power line channels. If the initially segmented power line channels consist of multiple parallel lines, these parallel lines are further segmented into multiple power line channels, achieving adaptive and correct segmentation of multiple power line channels in the scene. This facilitates the UAV's flight along designated power line channels during flight, achieving automated data acquisition of the power line channels.

[0080] To achieve the above objectives, see [link to relevant documentation]. Figure 1 , Figure 1 This is a flowchart illustrating an adaptive segmentation method for the line channel of an unmanned aerial vehicle (UAV)-borne lidar, provided in an embodiment of the present invention. Figure 1 As shown, the adaptive segmentation method for the line channel of the UAV-borne lidar includes:

[0081] S110: During UAV flight, the point cloud data collected by the lidar is processed in real time to extract multiple power line point clouds. The point cloud data processing process is the conversion of the raw point cloud data obtained by the lidar into world coordinates. UAVs typically have GNSS data, which is used to process the raw point cloud data for the corresponding locations into the world coordinate system. The resulting lidar points represent all objects scanned by the lidar, including power lines, poles, ground, buildings, and trees.

[0082] As a preferred embodiment, such as Figure 2 As shown, the steps for extracting multiple power line point clouds from the point cloud data collected by the lidar in real time include:

[0083] S111: Use a lidar to acquire point cloud data of the target power line area. Typically, the UAV-borne lidar is vertically fixed on the UAV, with its acquisition direction along the line of view towards the target power line area on the ground. In this way, the lidar can acquire point cloud data of the target power line area, which includes the three-dimensional coordinates of the laser points in the target power line area.

[0084] S112: Solve the point cloud data according to the relationship between the lidar coordinate system and the world coordinate system to obtain the point cloud data in the world coordinate system. The lidar coordinate system is usually polar coordinates, while the world coordinate system can be obtained from GNSS data of the POS system carried by the UAV. By combining the transformation relationship between the lidar coordinate system and the world coordinate system, the point cloud data in the lidar coordinate system can be solved into the world coordinate system to obtain the point cloud data in the world coordinate system.

[0085] S113: Based on a predetermined point cloud feature extraction algorithm, multiple power line point clouds are extracted from point cloud data in the world coordinate system. The predetermined point cloud feature extraction algorithm includes estimating the density and distribution of point cloud blocks to extract point cloud features (such as point cloud density and distribution location), then decomposing these point cloud features using SVD to construct feature factors, thereby calculating whether the point cloud within a certain range represents existing objects, and extracting multiple power line point clouds from the point cloud data; alternatively, deep learning methods are used to perform machine learning on the point cloud data to determine the objects in the target power line region corresponding to the power line point clouds, thereby extracting multiple power line point clouds.

[0086] Figure 1 The adaptive segmentation method for power line channels of UAV-borne LiDAR provided in the illustrated embodiment further includes the following steps after extracting point clouds of multiple power lines:

[0087] S120: According to the predetermined line fitting algorithm, the line equation of the power line corresponding to each power line point cloud in the multiple power line point clouds is fitted respectively; the predetermined line fitting algorithm can be a RANSAC algorithm or other line extraction algorithm. Through the predetermined line fitting algorithm, the line equation of the power line corresponding to each power line point cloud can be fitted from the multiple power line point clouds and saved.

[0088] As a preferred embodiment, such as Figure 3 As shown, the steps described above for fitting the electric line equations corresponding to each electric line point cloud in multiple electric line point clouds according to a predetermined line fitting algorithm include:

[0089] S121: Use the RANSAC algorithm to select the coordinates of any two laser points in the point cloud of each power line.

[0090] S122: Calculate the fitted center line of the electric field point cloud using the coordinates of any two laser points.

[0091] S123: Based on the formula for calculating straight-line distance, the distances from other laser points in the electric field point cloud to the fitted center line are obtained respectively.

[0092] S124: Count the number of other laser points whose distance from the fitted center line is less than or equal to the predetermined shortest distance value.

[0093] S125: Recalculate the fitting center line of the electric field point cloud by selecting the coordinates of any two laser points, until the number of other laser points whose distance to the fitting center line is less than or equal to the predetermined shortest distance value is the largest. Then, use the recalculated fitting center line as the electric field line equation corresponding to the electric field point cloud.

[0094] The technical solution provided by this invention selects the coordinates of any two laser points in each power line point cloud using the RANSAC algorithm, and then calculates the fitted center line of the power line point cloud using the coordinates of these two laser points. In this way, the distance from other laser points in the power line point cloud to the fitted center line can be calculated according to the straight-line distance calculation formula. Then, the number of points whose distance is less than or equal to the predetermined shortest distance value is counted. This iterative calculation is performed, and when the number of points obtained after reselecting the fitted center line is the largest, the recalculated fitted center line can be accurately determined as the straight-line equation of the power line corresponding to the power line point cloud.

[0095] Figure 1 The adaptive segmentation method for power line channels of UAV-borne lidar provided in the illustrated embodiment further includes the following steps after fitting the power line straight line equation corresponding to the point cloud of each power line:

[0096] S130: Select any power line as a reference line using the power line equation. Use the reference line to correct the direction of other power lines. Then, cluster parallel power lines into the same power line group using a parallel line clustering algorithm. The parallel clustering algorithm selects one line from all the power lines as the reference line, compares the angles between other power lines and the reference line, and determines if they are parallel. If the two lines are parallel, add the other power lines to the power line group containing the reference line. This iterative process accurately obtains line groups in all directions, clustering parallel power lines into the same power line group.

[0097] Specifically, as a preferred embodiment, such as Figure 4 As shown, the steps described above—selecting any electric field line as a reference line using the electric field line equation, using the reference line to correct the direction of other electric field lines, and clustering mutually parallel electric field lines into the same electric field group using a parallel line clustering algorithm—specifically include:

[0098] S131: Calculate all the power line lines corresponding to the power line point cloud using the power line line equation, and select any power line line from all the power line lines as the reference line for the power line grouping.

[0099] S132: The angle between the electric field lines of other electric field lines and the reference line.

[0100] S133: Determine whether the angle between other power lines and the reference line is less than or equal to a predetermined angle threshold.

[0101] S134: If the included angle of the power lines is less than or equal to a predetermined included angle threshold, then determine that other power lines are parallel to the reference line and add the other power lines to the power line group where the reference line is located.

[0102] S135: If the included angle of the electric field lines is greater than a predetermined included angle threshold, then select any electric field line from all electric field lines whose included angle is greater than the predetermined included angle threshold as a reference line, and repeat the steps of calculating the included angles of other electric field lines with the reference line and determining whether the included angles of other electric field lines with the reference line are less than or equal to the predetermined included angle threshold, until all parallel electric field lines are clustered into the same electric field group.

[0103] The main process of parallel line clustering in the technical solution provided by the embodiments of the present invention includes:

[0104] Step 1: Select one straight power line from all the power lines as a reference line;

[0105] Step 2: Select one line from the remaining lines as the current line. Use the direction of the reference line as the positive direction to correct the direction of the current line. Calculate the angle between the two lines. If the angle is greater than 90 degrees, the current line is considered to be in the opposite direction to the reference line. Reverse the direction of the current line to obtain a line in the same direction as the reference line and the angle between the two lines. This ensures that the angle between the current line and the reference line is less than 90 degrees.

[0106] Step 3: Determine whether the angle between the current line and the reference line exceeds a predetermined angle threshold. If it is within the threshold range, determine that the current line and the reference line are parallel, and add the current line to the group of reference lines to complete the line clustering process. Otherwise, continue to select power line lines from all remaining power line lines for comparison and repeat the above process.

[0107] Step 4: The above gives us a set of lines parallel to the reference line. If there are no lines among the remaining power line lines, then we only get one set of line channel directions. Otherwise, we continue to traverse all the remaining power line lines and repeat steps 1 to 3 until all power line lines in all directions are grouped.

[0108] Figure 1 The UAV-borne LiDAR line channel adaptive segmentation method provided in the illustrated embodiment, after the step of clustering mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm, further includes:

[0109] S140: Based on the line spacing of straight power lines, the same power line group is divided into multiple line sets. When dividing and aggregating parallel lines in a power line group, if multiple line corridors exist, the power line group is divided into multiple line corridors. Because each level of power line has prescribed design specifications, and its line spacing and corridor spacing have corresponding requirements, the corridor spacing relationship can be automatically estimated based on the line spacing within the corridor, thereby dividing the line corridors. That is, based on the line spacing of straight power lines, the corridor spacing is determined, and the power line group is further divided into multiple line sets. When the straight power lines in the power line group are completely divided, one line set corresponds to one line corridor.

[0110] S150: For each of the multiple line sets, calculate the minimum line spacing between the remaining straight power lines in the same power line group and the line set. Based on the relationship between the minimum line spacing and the average line spacing of the line set's corridors, the remaining straight power lines are assigned to the line sets. Because each level of power line has prescribed design specifications, and its line spacing and corridor spacing have corresponding requirements, this application automatically estimates the corridor spacing relationship based on the relationship between the minimum line spacing and the average line spacing of the line set's corridors, and assigns the remaining straight power lines in the power line groups to the line sets, thus adaptively segmenting the line corridors.

[0111] As a preferred embodiment, such as Figure 5 As shown, in the above adaptive segmentation method for power line channels, step S150: for each power line set in multiple power line sets, calculating the minimum line spacing between the remaining power line straight lines in the same power line group and the power line set, and segmenting the remaining power line straight lines into the power line set according to the relationship between the minimum line spacing and the average line spacing of the power line channel in the power line set, includes:

[0112] S151: Select a power line straight line from the power line group and add it to the currently segmented line set;

[0113] S152: Calculate the minimum line spacing between each remaining power line straight line and the line set in the same power line group.

[0114] In one preferred embodiment, such as Figure 6 As shown, in the above adaptive segmentation method for power line channels, step S152, which calculates the minimum line spacing between each remaining power line straight line and the line set in the same power line group, includes:

[0115] S1521: The straight lines of the power lines selected from the power line groups are used as the initial straight lines of the line sets; each line set selects one straight line from the aforementioned power line groups as a reference straight line to calculate the distance between the remaining straight lines of the aforementioned power line groups and the straight lines in the line set.

[0116] S1522: Calculate the distance between each remaining power line line in the same power line group and the initial line of the line set, as the minimum line spacing; or,

[0117] S1523: Calculate the distances between each remaining power line line in the same power line group and all lines in the line set;

[0118] S1524: Select the minimum distance between each remaining power line and all lines in the line set as the minimum line spacing;

[0119] S1525: When the spacing between the split lines is greater than the preset maximum spacing between the lines, determine whether there are any remaining straight power lines in the same power line group;

[0120] S1526: If there are remaining straight power lines in the same power line group, then all remaining straight power lines in the same power line group shall be merged into an independent line channel.

[0121] By selecting straight power lines from the power line group as the initial straight lines of the line set, and then calculating the distance between each remaining power line straight line and the initial straight line of the line set as the minimum line spacing, or by calculating the distances between each remaining power line straight line in the unified power line group and all straight lines in the line set, and then selecting the minimum distance as the minimum line spacing, all power lines in the power line group are added to the corresponding line set. All unadded power line straight lines are merged into independent line channels. The straight lines in the remaining line set P are often cables that are far from the line channel, and therefore are classified into independent line channels.

[0122] S153: Determine whether the minimum line spacing between each remaining power line straight line in the same power line group and the line set is less than the dividing line spacing. Let the minimum line spacing be d and the dividing line spacing be r. Then it is necessary to determine whether the minimum line spacing d is less than the dividing line spacing r, so as to avoid classifying power line straight lines that are far apart into the same line set.

[0123] S154: If the minimum line spacing is less than the segmented line spacing, then determine whether the minimum line spacing is less than the product of the average line spacing of the channel and the channel spacing factor of the line set. Let the minimum line spacing be d, the average line spacing of the road in the line set be D, and the channel spacing factor be s. Then, determine whether the minimum line spacing is less than the product of the average line spacing of the channel and the channel spacing factor, that is, determine whether d... <D*s。

[0124] S155: If the minimum line spacing is less than the product of the average line spacing and the channel spacing factor in the line set, then add the remaining power line straight lines to the line set.

[0125] S156: If the minimum line spacing is greater than or equal to the product of the average line spacing and the channel spacing factor in the line set, then the remaining power line straight lines are retained to the same power line group.

[0126] S157: Increment the spacing between the split lines and determine whether the spacing between the split lines is greater than the preset maximum spacing between the lines; if yes, proceed to step S158; if no, re-execute the step of determining whether the minimum line spacing between each remaining power line straight line and the line set in the same power line group is less than the spacing between the split lines, until the spacing between the split lines is greater than the preset maximum spacing between the lines.

[0127] S158: If so, when all power lines in the same power line group have been straightened out, all power lines in the same line set will be aggregated into the same line channel.

[0128] To adaptively extract adjacent power line lines, a maximum line spacing of dmax is set, and the initial line spacing is r (an initial r is a very small value to accurately divide line spacing and line channels). The single spacing increment step is t, and the line channel spacing factor is s. This means the minimum spacing between two channels should be greater than the product of the current average line spacing D and the channel spacing factor; otherwise, power line lines with a minimum spacing less than this product are considered to be in the same channel. Here, the remaining line set is labeled P, the current line set to be segmented is labeled Q, and each segmented set is labeled Qi. Previous segmentation methods simply placed power line lines in the same direction into the same set (segmented by an angle threshold). Further segmentation is to segment parallel lines, i.e., further subdividing by a distance threshold. The calculation method for the line spacing within each set Qi is as follows: first, calculate the shortest distance between each line in the set and its adjacent lines; then, count these shortest distances and take the maximum value of the shortest distance as the line spacing within the set. The process stops when the shortest spacing of the segmented line reaches dmax, and the line spacing within the set is no longer calculated.

[0129] The specific segmentation steps are as follows:

[0130] Step 1: Add all the power line straight lines in the deduplicated unified line group to the remaining line set P;

[0131] Step 2: Select a straight line from the remaining set of straight lines P as the initial straight line for the line set Qi, and add the initial straight line to the set. Regarding the selection of the initial straight line, according to the RANSAC detection results, lines with more point clouds appear earlier, so the initial straight line is the one with the most point clouds. Additionally, as a special case, when there is only one line in the set, the average spacing within the set is 0; in this case, there is only one straight line in Qi, which is the line in question.

[0132] Step 3: Select one line from the remaining line set P as the current line. Calculate the line spacing between the current line and all lines in the line set Qi in Q. Let the minimum line spacing be d. If the minimum line spacing d is less than the initial line spacing r, calculate the average line spacing D in the line set Qi. If the minimum line spacing d < the product of the average line spacing and the channel spacing factor s, i.e., D*s, then the current line can be assigned to the same line channel as the Qi set, and the current line is added to the line set Qi. If d > r, proceed to the next step.

[0133] Step 4: Repeat step 3 until all lines in the remaining set of lines have been traversed.

[0134] Step 5: Adjust the segmentation spacing r, r = r + t. After increasing r, continue executing steps 2 through 4.

[0135] If the final line spacing is greater than the maximum line spacing r>dmax, it means that the current line spacing has reached its maximum, and no further segmentation is performed. Instead, all lines in the remaining line set P are divided into independent line channels. These lines are often cables that are far from the line channels.

[0136] Step 6: Perform steps 1 to 6 for each group to achieve automatic channel segmentation for each group.

[0137] The above steps gradually increase the segmentation distance, from small to large, to achieve aggregation of small-pitch line channels and then to aggregation of large-pitch line channels. Simultaneously, line merging is performed based on the spacing characteristics of each channel.

[0138] Figure 1 The adaptive segmentation method for power line channels of UAV-borne LiDAR provided in the illustrated embodiment further includes, after segmenting the straight power lines in the same power line group:

[0139] S160: When all power line straight lines in the same power line group have been separated, all power line straight lines in the same line set will be aggregated into the same line channel.

[0140] In summary, the adaptive segmentation method for UAV-borne LiDAR lines provided in this application extracts multiple power line point clouds from the point cloud data collected by the LiDAR during UAV flight. Then, based on a predetermined line fitting algorithm, the equations of the corresponding power line lines for each power line point cloud are fitted. These equations are then used to calculate the power line lines, facilitating their placement into the corresponding line channels. After obtaining the power line lines for each power line point cloud, a parallel line clustering algorithm is used to cluster parallel power line lines into the same power line group, thus completing the initial segmentation. Since the same power line group may include multiple line channels, the power line group is divided into multiple line sets based on the line spacing of the power line lines. For each line set, the minimum line spacing between the remaining power line lines in the same power line group and the line set is calculated. Based on the relationship between this minimum line spacing and the average line spacing of the line set's channels, the remaining power line lines are segmented into the corresponding line sets. Therefore, all power line lines in the same line set can be aggregated into the same line channel. The above method enables adaptive segmentation of multiple line channels, effectively dividing multiple channels below the UAV and facilitating UAV flight along designated line channels during flight, thus achieving automated data collection of the channels. This adaptive line channel segmentation method can handle conventional and parallel lines in the main network and distribution network, as well as complex parallel lines adjacent to the main network and distribution network.

[0141] In a preferred embodiment, after clustering parallel power line lines into the same power line group in step S130, it is necessary to perform deduplication on all power line lines within the same power line group to remove duplicate or split power line lines that are too close together. Thus, after deduplicating the same power line group containing parallel power line lines, a group of identical or split power lines can be obtained.

[0142] like Figure 7 As shown, the above-mentioned adaptive segmentation method for power line channels, after step S130: selecting any power line as a reference line using the power line equation, using the reference line to correct the direction of other power lines, and clustering mutually parallel power lines into the same power line group according to the parallel line clustering algorithm, further includes: step S200: removing duplicate or split power lines in the same power line group based on the distance between power lines in the same power line group; Figure 7 As shown, the step of removing duplicate or split power line lines within the same power line group based on the distance between them includes:

[0143] S210: Select any straight line of the power line in the same power line group as the reference line of the container; specifically, construct a queue, store the first straight line of the current power line group into the queue, and then record the straight lines in the queue as reference lines into the unique corresponding container.

[0144] S220: Calculate the distance between all other power line lines in the same power line group and the reference line. This requires traversing all remaining power line lines in the current power line group, calculating the distance between all other power line lines and the reference line, and then determining whether the distance is less than or equal to the predetermined line distance.

[0145] S230: Determine whether the distance between each other power line and the reference line is less than or equal to the predetermined line distance.

[0146] S240: If the spacing is less than or equal to the predetermined straight line distance, then other power lines are identified as power line lines that repeat or split from the reference line, and the other power line lines are added to the container corresponding to the reference line.

[0147] S250: If the spacing is greater than the predetermined straight line distance, select any one of the power line lines with a spacing greater than the predetermined straight line distance as the reference line of the container, and repeat the step of determining whether the spacing between other power line lines and the reference line is less than or equal to the predetermined straight line distance, until all power line lines in the same power line group are added to the container of the corresponding reference line, wherein each reference line corresponds to a unique container.

[0148] Here, after traversing all remaining power line lines in the current power line group, the distance between each remaining power line line and the reference line is calculated. If the distance between the two lines is less than the predetermined line distance, it means that the distance between the two lines is very close, and they are regarded as duplicate or split conductors and added to the same queue. Conversely, if the distance exceeds the threshold of the predetermined line distance, the remaining power line lines are not processed.

[0149] If the queue is not empty, continue repeating the above steps until all nearby power lines are saved to the container of the corresponding reference line. If the remaining lines are not empty, continue removing lines from the remaining lines as reference lines, repeating the above steps, and saving the lines to a new container.

[0150] S260: Traverse all the power lines in all containers, select the highest power line in each container and merge them to obtain a deduplicated group of the same power line.

[0151] In the technical solution provided in this application embodiment, it is necessary to deduplicate all the same power line groups. Due to the quality issues of real-time point clouds, a single line may be identified as multiple lines. In addition, there are cases of split conductors in actual scenarios, and the above factors will affect the segmentation of the lines. To avoid the above problems, it is necessary to remove parallel and closely spaced lines in the same power line group. Among multiple adjacent and parallel power line straight lines, only the power line straight line with the highest height is retained to obtain the deduplicated lines.

[0152] As a preferred embodiment, such as Figure 8 As shown, the above-mentioned adaptive segmentation method for power line channels further includes, after step S130: clustering mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm:

[0153] S310: Determine whether there are multiple line channels in the same power line group based on the line spacing of the power line straight line;

[0154] S320: If there are multiple line channels in the same power line group, then perform the step of dividing the same power line group into multiple line sets according to the line spacing of the power line straight line;

[0155] S330: If there are no multiple line channels within the same power line group, then all power line lines in the same power line group will be aggregated into the same line channel.

[0156] In the technical solution provided by the embodiments of the present invention, parallel power line lines within the same power line group are divided and aggregated; the spacing between the divided power line lines is used to determine whether there are multiple line channels. If multiple line channels exist, all power line lines within the same power line group are automatically divided into multiple line channels according to the spacing between the divided line lines; otherwise, if there is only one line channel within the same power line group, all power line lines within the group are aggregated into the same line channel, thus achieving automatic segmentation of all lines.

[0157] In addition, based on the same concept of the above method embodiments, the present invention also provides an adaptive segmentation system for the line channel of UAV-borne lidar, which is used to implement the above method of the present invention. Since the principle and method of solving the problem in this system embodiment are similar, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, and will not be described in detail here.

[0158] See Figure 9 , Figure 9 This is a schematic diagram of the adaptive segmentation system for the line channel of an unmanned aerial vehicle (UAV)-borne lidar, provided by the present invention. (See diagram below.) Figure 9 As shown, the UAV-borne lidar line channel adaptive segmentation system includes:

[0159] The power line point cloud extraction module 110 is used to process the point cloud data collected by the lidar in real time during the flight of the UAV and extract multiple power line point clouds.

[0160] The power line straight line equation fitting module 120 is used to fit the power line straight line equation corresponding to each power line point cloud in multiple power line point clouds according to a predetermined line fitting algorithm.

[0161] The power line straight line clustering module 130 is used to select any power line straight line as a reference line using the power line straight line equation, use the reference line to correct the direction of other power line straight lines, and cluster mutually parallel power line straight lines into the same power line group according to the parallel line clustering algorithm.

[0162] The line set segmentation module 140 is used to group the same power line into multiple line sets according to the line spacing of the power line straight line;

[0163] The minimum line spacing calculation module 150 is used to calculate the minimum line spacing between the remaining power line straight lines and the line set in the same power line group for each of multiple line sets.

[0164] The power line straight line segmentation module 160 is used to segment the remaining power lines into the line set according to the relationship between the minimum line spacing and the average line spacing of the channel in the line set;

[0165] The power line straight line clustering module 170 is used to aggregate all power line straight lines of the same line set into the same line channel after all power line straight lines in the same power line group have been segmented.

[0166] In summary, the adaptive segmentation method for UAV-borne LiDAR lines provided in this application uses a power line point cloud extraction module 110 to solve the point cloud data collected by the LiDAR in real time during UAV flight, extracting multiple power line point clouds from the point cloud data. Then, a power line straight line equation fitting module 120 fits the power line straight line equation corresponding to each power line point cloud according to a predetermined line fitting algorithm. The power line straight line clustering module 130 then uses the power line straight line equation to calculate the power line straight line, facilitating the classification of the power line straight line into the corresponding line channel. After obtaining the power line straight line corresponding to each power line point cloud, a parallel line clustering algorithm is used to cluster mutually parallel power line straight lines into the same power line group, thereby completing the initial segmentation of the power line straight lines. Because the same power line group may include multiple line channels, the line set segmentation module 140 divides the power line group into multiple line sets based on the segmentation line spacing of the power line straight lines. The minimum line spacing calculation module 150 calculates the minimum line spacing between the remaining power line straight lines in the same power line group and the line set for each line set. The power line straight line segmentation module 160 segments the remaining power line straight lines into the corresponding line sets based on the relationship between the minimum line spacing and the average line spacing of the line set's channels. Then, the power line straight line clustering module 170 can aggregate all power line straight lines of the same line set into the same line channel. Through the above method, adaptive segmentation of multiple line channels can be achieved, effectively and correctly segmenting multiple channels below the UAV, facilitating the UAV to fly along the designated line channels during flight and realizing automated data collection of the channels. This adaptive line channel segmentation method can handle conventional lines and parallel lines in the main grid and distribution network, and can also handle complex parallel lines adjacent to the main grid and distribution network.

[0167] See Figure 10 , Figure 10 This is a schematic diagram of a line channel adaptive segmentation system for an unmanned aerial vehicle (UAV)-borne lidar, provided as an embodiment of the present invention. Figure 10 As shown, the line channel adaptive segmentation system includes: a processor 1001, a communication bus 1002, a communication module 1003, a memory 1004, and a line channel adaptive segmentation program for an UAV-borne LiDAR stored in the memory 1004 and running on the processor 1001. When the tracking channel adaptive segmentation program is executed by the processor, it implements the steps of the line channel adaptive segmentation method provided in any of the above embodiments.

[0168] In summary, compared with existing technologies, the adaptive line channel segmentation scheme for UAV-borne LiDAR provided in the above embodiments of the present invention offers a reliable and adaptive line segmentation algorithm that can effectively and correctly segment multiple channels below the UAV, facilitating the UAV's flight along designated channels and enabling automated channel data acquisition. This method can handle conventional and parallel lines in the main network and distribution network, as well as complex parallel lines adjacent to each other in the main network and distribution network.

[0169] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0170] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0171] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0172] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0173] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0174] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0175] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An adaptive segmentation method for the line channel of an unmanned aerial vehicle (UAV)-borne lidar, characterized in that, include: During the drone's flight, the point cloud data collected by the lidar is processed in real time to extract multiple power line point clouds; According to the predetermined line fitting algorithm, the equation of the power line line corresponding to each power line point cloud in the multiple power line point clouds is fitted respectively; The electric field line equation is used to select any electric field line as a reference line. The reference line is used to correct the direction of other electric field lines. Parallel electric field lines are clustered into the same electric field group according to the parallel line clustering algorithm. Based on the spacing between the straight power lines, the same power line is divided into multiple line sets. For each of the multiple line sets, calculate the minimum line spacing between the remaining power line straight lines in the same power line group and the line set. Based on the relationship between the minimum line spacing and the average line spacing of the channel in the line set, divide the remaining power line straight lines into the line set. When all power line lines in the same power line group have been separated, all power line lines in the same line set will be aggregated into the same line channel. The step of calculating the minimum line spacing between the remaining power line straight lines in the same power line group and the line set for each of the plurality of line sets, and dividing the remaining power line straight lines into the line set according to the relationship between the minimum line spacing and the average line spacing of the channel in the line set, includes: Select a straight power line from the power line group and add it to the currently segmented line set; Calculate the minimum line spacing between each remaining power line straight line in the same power line group and the line set; Determine whether the minimum line spacing between each remaining power line straight line in the same power line group and the line set is less than the line spacing of the segmented lines; If the minimum line spacing is less than the segmented line spacing, then determine whether the minimum line spacing is less than the product of the channel average line spacing of the line set and the channel spacing factor. If the minimum line spacing is less than the product of the average line spacing and the channel spacing factor in the line set, then the remaining power line straight lines are added to the line set. If the minimum line spacing is greater than or equal to the product of the average line spacing and the channel spacing factor in the line set, then the remaining power line straight lines are retained to the same power line group. Increment the spacing between the segmented lines and re-execute the step of determining whether the minimum line spacing between each remaining power line straight line in the same power line group and the line set is less than the spacing between the segmented lines, until the spacing between the segmented lines is greater than the preset maximum line spacing. When all power lines in the same power line group have been straightened out, all power lines in the same line set are aggregated into the same line channel.

2. The adaptive segmentation method for line channels according to claim 1, characterized in that, The step of extracting multiple power line point clouds from the point cloud data collected by the lidar in real time includes: The lidar is used to collect point cloud data of the target power line area; The point cloud data is calculated based on the relationship between the lidar coordinate system and the world coordinate system to obtain the point cloud data in the world coordinate system. According to the predetermined point cloud feature extraction algorithm, multiple power line point clouds are extracted from the point cloud data in the world coordinate system.

3. The adaptive segmentation method for line channels according to claim 1, characterized in that, The step of fitting the power line straight line equation corresponding to each power line point cloud in the plurality of power line point clouds according to the predetermined line fitting algorithm includes: The RANSAC algorithm is used to select the coordinates of any two laser points in the point cloud of each power line; The fitted center line of the electric field point cloud is obtained by calculating the coordinates of any two laser points. According to the formula for calculating straight-line distance, the distances from other laser points in the electric field point cloud to the fitted center line are obtained respectively; The number of points whose distance from the other laser points to the fitted center line is less than or equal to the predetermined shortest distance value is counted. Recalculate the fitted center line of the electric field point cloud by selecting the coordinates of any two laser points again, until the number of other laser points whose distance to the fitted center line is less than or equal to the predetermined shortest distance value is the largest. Then, take the recalculated fitted center line as the electric field line equation corresponding to the electric field point cloud.

4. The adaptive segmentation method for line channels according to claim 1, characterized in that, The steps of selecting any power line as a reference line using the power line equation, using the reference line to correct the direction of other power lines, and clustering mutually parallel power lines into the same power line group according to the parallel line clustering algorithm include: The electric field line straight line corresponding to the electric field line point cloud is calculated using the electric field line straight line equation. Select any one of the power line lines from all the power line lines as the reference line for the power line group, and calculate the power line angle between the other power line lines in the group and the reference line. Determine whether the angle between the other electric field lines and the reference line is less than or equal to a predetermined angle threshold. If the included angle of the power lines is less than or equal to a predetermined included angle threshold, then the other power line lines are determined to be parallel to the reference line, and the other power line lines are added to the power line group where the reference line is located; If the included angle of the electric field lines is greater than a predetermined included angle threshold, then any electric field line with an included angle greater than the predetermined included angle threshold is selected as a reference line, and the step of determining whether the included angle of the other electric field lines with the reference line is less than or equal to the predetermined included angle threshold is repeated until all parallel electric field lines are clustered into the same electric field line group.

5. The adaptive segmentation method for line channels according to claim 1, characterized in that, After the step of clustering mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm, the method further includes removing duplicate or split power line lines in the same power line group based on the distance between the power line lines in the same power line group; The step of removing duplicate or split power line lines in the same power line group based on the distance between power line lines in the same power line group includes: Select any straight line of the power line in the same power line group as the reference line of the container, and calculate the distance between all other straight lines of the power line in the same power line group and the reference line; Determine whether the distance between each of the other electric field lines and the reference line is less than or equal to a predetermined straight line distance; If the spacing is less than or equal to the predetermined straight line distance, then the other electric field line is determined to be an electric field line that repeats or splits from the reference line, and the other electric field line is added to the container corresponding to the reference line; If the spacing is greater than the predetermined straight line distance, then select any one of the power line lines with a spacing greater than the predetermined straight line distance as the reference line of the container, and repeat the step of determining whether the spacing between the other power line lines and the reference line is less than or equal to the predetermined straight line distance, until all power line lines in the same power line group are added to the container of the corresponding reference line, wherein each reference line corresponds to a unique container. Traverse all the power lines in all containers, select the highest power line from each container and merge them to obtain a group of identical power lines after deduplication.

6. The adaptive segmentation method for line channels according to claim 1, characterized in that, After the step of clustering mutually parallel power line lines into the same power line group according to the parallel line clustering algorithm, the method further includes: Based on the spacing between the dividing lines of the straight power lines, determine whether there are multiple line channels in the same power line group; If the same power line group has multiple line channels, then the step of dividing the same power line group into multiple line sets according to the line spacing of the power line straight line is performed; If there are no multiple line channels within the same power line group, then all power line lines in the same power line group will be aggregated into the same line channel.

7. The adaptive segmentation method for line channels according to claim 6, characterized in that, The step of calculating the minimum line spacing between each remaining power line straight line in the same power line group and the line set includes: The straight lines of the power lines selected from the power line groups are used as the initial straight lines of the line set; Calculate the distance between each remaining power line straight line in the same power line group and the initial straight line of the line set, and use this distance as the minimum line spacing; or, Calculate the distances between each remaining power line straight line in the same power line group and all straight lines in the line set; The minimum distance between each remaining power line and all the lines in the line set is selected as the minimum line spacing. When the spacing between the divided lines is greater than the preset maximum spacing between lines, it is determined whether there are any remaining straight power lines in the same power line group. If there are any remaining straight power lines in the same power line group, then all the remaining straight power lines in the same power line group will be merged into an independent line channel.

8. A line channel adaptive segmentation system for UAV-borne lidar, characterized in that, The method for adaptive segmentation of the line channel of an UAV-borne lidar according to any one of claims 1-7 includes: The power line point cloud extraction module is used to process the point cloud data collected by the lidar in real time during the flight of the UAV and extract multiple power line point clouds. The power line straight line equation fitting module is used to fit the power line straight line equation corresponding to each power line point cloud in the multiple power line point clouds according to a predetermined line fitting algorithm. The power line straight line clustering module is used to select any power line straight line as a reference line using the power line straight line equation, use the reference line straight line to correct the direction of other power line straight lines, and cluster mutually parallel power line straight lines into the same power line group according to the parallel line clustering algorithm. The line set segmentation module is used to group the same power line into multiple line sets according to the line spacing of the power line straight line; The minimum line spacing calculation module is used to calculate the minimum line spacing between the remaining power line straight lines in the same power line group and the line group for each of the multiple line groups; A power line straight-line segmentation module is used to segment the remaining power lines into the line set according to the relationship between the minimum line spacing and the average line spacing of the channel of the line set; The power line straight line clustering module is used to aggregate all power line straight lines of the same line set into the same line channel after all power line straight lines in the same power line group have been segmented.

9. A line channel adaptive segmentation system for an unmanned aerial vehicle (UAV)-borne lidar, characterized in that, include: The memory, the processor, and the UAV-borne LiDAR line channel adaptive segmentation program stored in the memory and running on the processor, wherein the line channel adaptive segmentation program, when executed by the processor, implements the steps of the line channel adaptive segmentation method as described in any one of claims 1 to 7.

Images