Transmission conductor spatial distance detection method based on laser radar and vision measurement, storage medium and equipment

Abstract

The invention discloses a transmission conductor spatial distance detection method based on laser radar and vision measurement, a storage medium and equipment, and relates to the technical field of electric power. The method specifically comprises the steps of S1, an online calibration algorithm based on the laser radar and the binocular camera and S2, a transmission conductor space distance detection algorithm based on fusion of the laser radar and the binocular camera. According to the scheme of combining the laser radar and binocular vision measurement, the laser radar and the camera are subjected to advantage complementation, so that the transmission conductor windage yaw waving track is accurately researched and judged, field personnel are assisted in accurately mastering the transmission conductor windage yaw waving amplitude information, the field monitoring precision and breadth are improved, and the working efficiency is improved. Active protection is changed into active control, and safe operation of power grid equipment is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power technology, specifically to a method, storage medium, and device for detecting spatial distances of power transmission lines based on lidar and visual measurement. Background Technology

[0002] As the carrier of electricity transmission, ensuring the safe and reliable operation of transmission lines is a crucial aspect of modern power grid construction. Since transmission lines are erected in the natural environment, their structural safety and stability are easily affected by environmental factors. Galloping is one of the common types of faults in transmission lines. Under certain angles of attack and wind speeds, conductors are prone to large-amplitude, low-frequency self-excited vibrations, which is called galloping. When a transmission line fault occurs, it not only affects the local power supply but also often causes grid oscillations and damage to the grid, leading to large-scale power outages. Research on transmission line galloping monitoring helps in the timely development of disaster response plans, ensuring the stable operation of the power supply system and safeguarding the safety of electricity use in society.

[0003] There are three main technologies for monitoring power transmission line galloping: monitoring using video images, monitoring using differential GPS to obtain the three-dimensional coordinates of the transmission line to be monitored, and analyzing the galloping situation by collecting conductor galloping parameters through sensors. Existing technologies cannot provide quantitative monitoring using video monitoring, using differential GPS is costly, and sensor monitoring can only reflect the situation at a single point on the conductor and is difficult to place. Vision-based perception systems achieve satisfactory performance at low cost, but single-camera perception systems cannot provide reliable 3D geometry. On the other hand, stereo cameras can provide 3D geometry, but in practical applications, they are costly and difficult to use efficiently in heavily obstructed and textureless environments. Furthermore, camera-based perception systems are difficult to operate under complex or poor lighting conditions, limiting their all-weather capability. LiDAR can provide high-precision 3D geometry and is unaffected by ambient light. However, mobile LiDAR is limited by low resolution (from 16 to 128 channels) and low refresh rate (10Hz). Summary of the Invention

[0004] The purpose of this invention is to provide a method for detecting the spatial distance of power transmission lines based on lidar and visual measurement, so as to solve the problems mentioned in the background art that the existing spatial distance detection technology for power transmission lines cannot accurately identify the range of galloping and cannot be detected at night.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for spatial distance detection of power transmission lines based on lidar and visual measurement, which detects power transmission lines using lidar and a binocular camera, specifically including:

[0006] S1 is an online calibration algorithm based on LiDAR and binocular camera: it performs coarse registration of the image and removes abnormal features, and after removing abnormal features, it performs fine registration of LiDAR and camera and outputs data.

[0007] S2 is a spatial distance detection algorithm for power transmission lines based on the fusion of LiDAR and binocular camera: The data output by S1 is fused with LiDAR and binocular camera to achieve high-precision extraction and modeling of power transmission lines.

[0008] Preferably, S1 is based on an online calibration algorithm using LiDAR and a binocular camera, specifically as follows:

[0009] S11: Enhance image contrast using histogram equalization and perform coarse registration based on the SuperGlue algorithm;

[0010] S12: Use the RANSAC algorithm to remove outdated feature matching pairs from the data output by S11;

[0011] S13: After removing abnormal features, a NID-based method is used to perform fine registration between the LiDAR and the camera to achieve more accurate spatial alignment.

[0012] S14: Consider the real-time performance of the algorithm and integrate the optimized scheme into the actual system to provide a reliable spatial extrinsic parameter solution technology for power line spatial distance monitoring technology, so as to obtain and output accurate fused data.

[0013] Preferably, S2 is a spatial distance detection algorithm for power transmission lines based on the fusion of lidar and binocular camera, specifically as follows:

[0014] S21: Utilizing image-based power transmission line edge extraction technology, initial values ​​and visualization effects are provided for subsequent point cloud-based power transmission line extraction algorithms;

[0015] S22: Based on S21, the LK optical flow method is used to register the acquired lidar point cloud data, construct an incremental map, and extract the initial power transmission line point cloud using spatial constraint characteristics.

[0016] S23: Further optimize the point cloud extraction algorithm for transmission lines to achieve accurate extraction, stranding, and fitting of transmission lines, ultimately obtaining a complete three-dimensional model of the transmission lines.

[0017] Preferably, S21, the image-based transmission line edge extraction technology, specifically includes the following:

[0018] S211 Image Enhancement: Enhances the original image, including grayscale conversion, histogram equalization, and grayscale transformation;

[0019] S212 Improved Ratio Edge Detection Algorithm: An improved Ratio edge detection algorithm is used to extract the edges of power lines and remove complex vegetation background information that affects the edges of power lines after image enhancement processing.

[0020] S213 Hough Transform: After processing with the improved Ratio edge detection algorithm, the Hough transform is used to extract power transmission lines. The method of mapping the edge points in the image to the parameter space and finding collinear points in the parameter space realizes the detection and extraction of straight lines.

[0021] Preferably, the grayscale processing, histogram equalization, and grayscale transformation in image enhancement S211 are as follows:

[0022] Grayscale processing: The acquired power transmission lines are processed to obtain grayscale images; statistical analysis is performed on the grayscale images of the power lines to obtain the grayscale image histogram;

[0023] Histogram equalization: Adjusts the distribution of each gray level in the image to make the distribution of each gray level more uniform, so that the gray level image no longer exhibits a single peak, and enhances the feature information of the power transmission line in the image.

[0024] Gray-scale transformation: By defining a gray-scale transformation function, gray-scale mapping is performed on each pixel of the power transmission line image to enhance the contrast and clarity of the power transmission line in the image;

[0025] The S212 improved Ratio edge detection algorithm uses an improved Ratio edge detection algorithm to extract the edges of power lines, as follows:

[0026] S2121 in the image Selected from Set the middle neighborhood as the center pixel. Left Neighbor and right field ;

[0027] S2122 Calculate the neighborhood respectively , , The average grayscale value is calculated using the following formula:

[0028]

[0029] In the formula, This is the average grayscale value. This represents the total number of pixels in the neighborhood. The number of image points in the neighborhood. ;

[0030] S2123 defines the edge response function. It is expressed by the following formula:

[0031]

[0032] In the formula and The grayscale mean is... , ;

[0033] S2124 Request response function value at And store in an array It is expressed by the following formula:

[0034]

[0035] Repeat steps S2121-S2124, moving... Iterate through all positions and find the maximum array. It is expressed by the following formula:

[0036]

[0037] S2125 sets the line feature detection threshold. The reference range is The edge detection result is It is expressed by the following formula:

[0038] .

[0039] Preferably, S22 uses the LK optical flow method to register the acquired lidar point cloud data, specifically including the following:

[0040] S221 Spatial Feature Constraint Filtering: Spatial feature constraint filtering divides the acquired point cloud data into grids and calculates the elevation difference between the highest and lowest elevation points in the grid.

[0041] If the elevation difference is less than the elevation threshold, the point in the grid is initially determined to be a ground feature point, and further statistical analysis is performed on the grid point cloud that is greater than the elevation threshold.

[0042] If the elevation distribution of the point cloud within the grid shows multiple peaks and there is a continuous gap in the elevation values, it can be determined that the grid contains the point cloud of the power transmission line.

[0043] S222 Power Line Extraction: The idea of ​​Hough transform to detect straight lines is to map a straight line in the original coordinate space to a point in the parameter space through coordinate transformation; when the curves intersect at multiple points, it is judged that the intersection point indicates that these points show obvious linear characteristics. Therefore, the more prominent the accumulated value is in the local range, the more the corresponding parameter matches the mathematical expression of the straight line in the original space.

[0044] S223 power line splitting, specifically including:

[0045] S2231: Select a point cloud as a seed point from the complete power line data obtained after the fine extraction operation;

[0046] S2232: Traverse all laser point clouds except the seed point and calculate the distance between each point cloud and the seed point. Find its nearest neighbor in terms of spatial distance and record its number. ;

[0047] S2233: Set a distance threshold If the distance is obtained according to S2232 Less than Then the number is Place the point and the seed point together; then The point is used as the new seed point, and the remaining points are traversed.

[0048] S2234: When the distances between the remaining points and the seed point do not meet the distance threshold. This indicates that the identification of a complete power line point cloud has been completed.

[0049] S2235: After identifying a power line, randomly select one point from the remaining points as a new seed point, and repeat S2232, S2233 and S2234 until no new point can be found as a seed point; complete the identification and separation of each power line in the complete power line system.

[0050] Preferably, the S221 spatial feature constraint filtering specifically includes:

[0051] Transform the point cloud of ground features into a 3D grid and then convert the voxels containing the points into a grid. Marked as 1;

[0052] All voxels with the same position on the plane are considered as the basic unit of judgment, i.e. directional cubic pillars, and the size of the interval of statistical blank voxels based on the presence or absence of data points;

[0053] For a single column If there is only one layer of point cloud, then this cubic column does not contain power line point cloud; if there are multiple layers of points and the interlayer interval is less than the interval threshold finterval, then this column does not contain power line points; if there are multiple layers of points and the interlayer interval is greater than the interval threshold finterval, then the points below the layer with the largest interlayer interval are removed, and the remaining points are candidate power line points.

[0054] Preferably, S23 specifically involves: further optimizing the point cloud extraction algorithm for transmission lines, and using a feature fusion-based method to fuse features from the image and point cloud data; achieving accurate extraction, strand division, and fitting of the transmission lines, ultimately obtaining a complete three-dimensional model of the transmission lines.

[0055] An electronically readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, performs the aforementioned method for detecting the spatial distance of power transmission lines based on lidar and visual measurement.

[0056] An electronic device, the electronic device comprising:

[0057] processor;

[0058] The memory stores a computer program that, when executed by the processor, performs the steps of the above-described method for detecting the spatial distance of power transmission lines based on lidar and visual measurement.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] This invention combines lidar and binocular vision measurement, leveraging the complementary advantages of these two sensors to accurately assess the wind-induced galloping trajectory of power transmission lines. This assists on-site personnel in precisely understanding the amplitude of wind-induced galloping, improving the accuracy and scope of on-site monitoring, transforming proactive protection into proactive control, and ensuring the safe operation of power grid equipment. Attached Figure Description

[0061] The accompanying drawings, as part of this invention, are provided to further illustrate the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention, but do not constitute an undue limitation thereof. Clearly, the drawings described below are merely some embodiments, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0062] Figure 1 This is a flowchart of the online calibration algorithm based on lidar and binocular camera for S1 of the present invention, which is a method for detecting spatial distance of power transmission lines based on lidar and vision measurement.

[0063] Figure 2 S2 is a flowchart of the spatial distance detection algorithm for power transmission lines based on the fusion of lidar and binocular camera, which is a method for spatial distance detection of power transmission lines based on lidar and vision measurement according to the present invention.

[0064] Figure 3 This is a flowchart of a method for detecting spatial distance of power transmission lines based on lidar and visual measurement according to the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Please see Figures 1-3 A method for spatial distance detection of power transmission lines based on lidar and visual measurement, such as... Figure 3 As shown, the transmission lines are inspected using a lidar and a binocular camera, specifically including:

[0067] S1 is an online calibration algorithm based on LiDAR and binocular camera: it performs coarse registration of the image and removes abnormal features, and after removing abnormal features, it performs fine registration of LiDAR and camera and outputs data.

[0068] S2 is a spatial distance detection algorithm for power transmission lines based on the fusion of LiDAR and binocular camera: The data output by S1 is fused with LiDAR and binocular camera to achieve high-precision extraction and modeling of power transmission lines.

[0069] As further explanation:

[0070] like Figure 1 As shown, S1 is based on the online calibration algorithm of LiDAR and binocular camera, specifically as follows:

[0071] S11: Enhance image contrast using histogram equalization and perform coarse registration based on the SuperGlue algorithm.

[0072] S12: Use the RANSAC algorithm to remove outlier feature matching pairs from the data output by S11.

[0073] S13: After removing abnormal features, a NID-based method is used to perform fine registration between the LiDAR and the camera to achieve more accurate spatial alignment.

[0074] S14: Consider the real-time performance of the algorithm and integrate the optimized scheme into the actual system to ensure stability and reliability in the real-time environment. Provide a reliable spatial extrinsic parameter solution technology for power line spatial distance monitoring technology to obtain and output accurate fused data.

[0075] like Figure 2 As shown, the 2S2 algorithm for spatial distance detection of power transmission lines based on the fusion of lidar and binocular cameras is as follows:

[0076] S21: Utilizing image-based power transmission line edge extraction technology, initial values ​​and visualization effects are provided for subsequent point cloud-based power transmission line extraction algorithms.

[0077] The image-based power transmission line edge extraction technology specifically includes the following:

[0078] S211 Image Enhancement: Due to the complex background of the images acquired by the monitoring platform, it is difficult to directly extract the power transmission lines from the original images. Therefore, enhancement processing is performed on the original images, including grayscale conversion, histogram equalization, and grayscale transformation.

[0079] It should be noted that:

[0080] Grayscale processing: The acquired power transmission lines are processed to obtain grayscale images; statistical analysis is performed on the grayscale images of the power lines to obtain the grayscale image histogram.

[0081] Histogram equalization: Adjusts the distribution of each gray level in the image to make the distribution of each gray level more uniform, so that the gray image no longer exhibits a single peak, and enhances the feature information of the power transmission line in the image.

[0082] Gray-scale transformation: As an important technique in image processing, gray-scale transformation plays a crucial role in the detection and analysis of power transmission lines. By defining a gray-scale transformation function, each pixel in the transmission line image is mapped to a gray level, enhancing the contrast and clarity of the transmission lines in the image, thereby improving the accuracy and reliability of line detection.

[0083] The S212 improved Ratio edge detection algorithm: This algorithm extracts the edges of power lines by removing complex vegetation background information, such as trees and weeds, that affects the edges of the power lines after image enhancement processing. Specifically, it works as follows:

[0084] S2121 in the image Selected from Set the middle neighborhood as the center pixel. Left Neighbor and right field ;

[0085] S2122 Calculate the neighborhood respectively , , The average grayscale value is calculated using the following formula:

[0086]

[0087] In the formula, This is the average grayscale value. This represents the total number of pixels in the neighborhood. The number of image points in the neighborhood. ;

[0088] S2123 defines the edge response function. It is expressed by the following formula:

[0089]

[0090] In the formula and The grayscale mean is... , ;

[0091] S2124 Request response function value at And store in an array It is expressed by the following formula:

[0092]

[0093] Repeat steps S2121-S2124, moving... Iterate through all positions and find the maximum array. It is expressed by the following formula:

[0094]

[0095] S2125 sets the line feature detection threshold. The reference range is The edge detection result is It is expressed by the following formula:

[0096] .

[0097] S213 Hough Transform: After processing with an improved Ratio edge detection algorithm, the Hough transform is used to extract power transmission lines. It detects and extracts straight lines by mapping edge points in the image to a parameter space and finding collinear points within that space. The Hough transform is an image processing technique that parameterizes each edge point in the image, forming a series of curves. Straight lines in the image are determined by finding the intersection points of these curves in the parameter space.

[0098] S22: Based on S21, the acquired lidar point cloud data is registered using the LK optical flow method to construct an incremental map, and the initial transmission line point cloud is extracted using spatial constraint characteristics; specifically including the following:

[0099] S221 Spatial Feature Constraint Filtering: For safety reasons, power transmission lines are installed at a certain height above the ground. Therefore, the spatial characteristics of the power transmission line point cloud are used to directly filter out most ground feature point clouds. Spatial feature constraint filtering divides the acquired point cloud data into a grid and calculates the elevation difference between the highest and lowest points in the grid. If this elevation difference is less than an elevation threshold, the points in that grid are initially identified as ground features. Further statistical analysis is performed on grid point clouds with elevation differences greater than the threshold. If the elevation distribution of the point cloud within a grid shows multiple peaks and there is a continuous gap in the elevation values, then it can be determined that the grid contains power transmission line point clouds.

[0100] The specific operation process of the above spatial feature constraint filtering includes:

[0101] Transform the point cloud of ground features into a 3D grid and then convert the voxels containing the points into a grid. Marked as 1;

[0102] All voxels with the same position on the plane are considered as the basic unit of judgment, i.e. directional cubic pillars, and the size of the interval of statistical blank voxels based on the presence or absence of data points;

[0103] For a single column If there is only one layer of point cloud, then this cubic column does not contain power line point cloud; if there are multiple layers of points and the interlayer interval is less than the interval threshold finterval, then this column does not contain power line points; if there are multiple layers of points and the interlayer interval is greater than the interval threshold finterval, then the points below the layer with the largest interlayer interval are removed, and the remaining points are candidate power line points.

[0104] S222 Power Line Extraction: The idea behind Hough transform for detecting straight lines is to map a straight line in the original coordinate space to a point in the parameter space through coordinate transformation. When multiple curves intersect at a single point, it is judged that the intersection point indicates that these points exhibit obvious linear characteristics. Therefore, the more prominent the accumulated value is in the local range, the more the corresponding parameters conform to the mathematical expression of the straight line in the original space.

[0105] S223 power line splitting, specifically including:

[0106] S2231: Select a point cloud as a seed point from the complete power line data obtained after the fine extraction operation;

[0107] S2232: Traverse all laser point clouds except the seed point and calculate the distance between each point cloud and the seed point. Find its nearest neighbor in terms of spatial distance and record its number. ;

[0108] S2233: Set a distance threshold If the distance is obtained according to S2232 Less than Then the number is Place the point and the seed point together; then The point is used as the new seed point, and the remaining points are traversed.

[0109] S2234: When the distances between the remaining points and the seed point do not meet the distance threshold. This indicates that the identification of a complete power line point cloud has been completed.

[0110] S2235: After identifying a power line, randomly select one point from the remaining points as a new seed point, and repeat S2232, S2233 and S2234 until no new point can be found as a seed point; complete the identification and separation of each power line in the complete power line system.

[0111] S23: Further optimize the point cloud extraction algorithm for transmission lines by using a feature fusion-based method to fuse features from the image and point cloud data; achieve accurate extraction, stranding, and fitting of transmission lines, and finally obtain a complete three-dimensional model of the transmission line.

[0112] An electronically readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, performs the aforementioned method for detecting the spatial distance of power transmission lines based on lidar and visual measurement.

[0113] An electronic device, the electronic device comprising:

[0114] processor;

[0115] The memory stores a computer program that, when executed by the processor, performs the steps of the above-described method for detecting the spatial distance of power transmission lines based on lidar and visual measurement.

[0116] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting spatial distance of power transmission lines based on lidar and visual measurement, characterized in that, The detection of power transmission lines is carried out using lidar and binocular cameras, specifically including: S1 is an online calibration algorithm based on LiDAR and binocular camera: it performs coarse registration of the image and removes abnormal features, and after removing abnormal features, it performs fine registration of LiDAR and camera and outputs data. S2 is a spatial distance detection algorithm for power transmission lines based on the fusion of LiDAR and binocular camera: The data output by S1 is fused with LiDAR and binocular camera to achieve high-precision extraction and modeling of power transmission lines.

2. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 1, characterized in that: S1 is based on an online calibration algorithm using LiDAR and a binocular camera, as follows: S11: Enhance image contrast using histogram equalization and perform coarse registration based on the SuperGlue algorithm; S12: Use the RANSAC algorithm to remove outdated feature matching pairs from the data output by S11; S13: After removing abnormal features, a NID-based method is used to perform fine registration between the LiDAR and the camera to achieve more accurate spatial alignment. S14: Consider the real-time performance of the algorithm and integrate the optimized scheme into the actual system to provide a reliable spatial extrinsic parameter solution technology for power line spatial distance monitoring technology, so as to obtain and output accurate fused data.

3. The method for spatial distance detection of power transmission lines based on lidar and visual measurement according to claim 1 or 2, characterized in that: S2 is a spatial distance detection algorithm for power transmission lines based on the fusion of LiDAR and binocular cameras, as detailed below: S21: Utilizing image-based power transmission line edge extraction technology, initial values ​​and visualization effects are provided for subsequent point cloud-based power transmission line extraction algorithms; S22: Based on S21, the LK optical flow method is used to register the acquired lidar point cloud data, construct an incremental map, and extract the initial power transmission line point cloud using spatial constraint characteristics. S23: Further optimize the point cloud extraction algorithm for transmission lines to achieve accurate extraction, stranding, and fitting of transmission lines, ultimately obtaining a complete three-dimensional model of the transmission lines.

4. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 3, characterized in that: S21 is an image-based power transmission line edge extraction technology, which specifically includes the following: S211 Image Enhancement: Enhances the original image, including grayscale conversion, histogram equalization, and grayscale transformation; S212 Improved Ratio Edge Detection Algorithm: An improved Ratio edge detection algorithm is used to extract the edges of power lines and remove complex vegetation background information that affects the edges of power lines after image enhancement processing. S213 Hough Transform: After processing with the improved Ratio edge detection algorithm, the Hough transform is used to extract power transmission lines. The method of mapping the edge points in the image to the parameter space and finding collinear points in the parameter space realizes the detection and extraction of straight lines.

5. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 4, characterized in that: The grayscale processing, histogram equalization, and grayscale transformation in S211 image enhancement are as follows: Grayscale processing: The acquired power transmission lines are processed to obtain grayscale images; statistical analysis is performed on the grayscale images of the power lines to obtain the grayscale image histogram; Histogram equalization: Adjusts the distribution of each gray level in the image to make the distribution of each gray level more uniform, so that the gray level image no longer exhibits a single peak, and enhances the feature information of the power transmission line in the image. Gray-scale transformation: By defining a gray-scale transformation function, gray-scale mapping is performed on each pixel of the power transmission line image to enhance the contrast and clarity of the power transmission line in the image; The S212 improved Ratio edge detection algorithm uses an improved Ratio edge detection algorithm to extract the edges of power lines, as follows: S2121 in the image Selected from Set the middle neighborhood as the center pixel. Left Neighbor and right field ; S2122 Calculate the neighborhood respectively , , The average grayscale value is calculated using the following formula: ； In the formula, This is the average grayscale value. The total number of pixels in the neighborhood. The number of image points in the neighborhood. ; S2123 defines the edge response function. It is expressed by the following formula: ； In the formula and The grayscale mean is... , ; S2124 Request response function value at And store in an array It is expressed by the following formula: ； Repeat steps S2121-S2124, moving... Iterate through all positions and find the maximum array. It is expressed by the following formula: ； S2125 sets the line feature detection threshold. The reference range is The edge detection result is It is expressed by the following formula: 。 6. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 3, characterized in that: S22 uses the LK optical flow method to register the acquired lidar point cloud data, specifically including the following: S221 Spatial Feature Constraint Filtering: Spatial feature constraint filtering divides the acquired point cloud data into grids and calculates the elevation difference between the highest and lowest elevation points in the grid. If the elevation difference is less than the elevation threshold, the point in the grid is initially determined to be a ground feature point, and further statistical analysis is performed on the grid point cloud that is greater than the elevation threshold. If the elevation distribution of the point cloud within the grid shows multiple peaks and there is a continuous gap in the elevation values, it can be determined that the grid contains the point cloud of the power transmission line. S222 Power Line Extraction: The idea of ​​Hough transform to detect straight lines is to map a straight line in the original coordinate space to a point in the parameter space through coordinate transformation; when the curves intersect at multiple points, it is judged that the intersection point indicates that these points show obvious linear characteristics. Therefore, the more prominent the accumulated value is in the local range, the more the corresponding parameter matches the mathematical expression of the straight line in the original space. S223 power line splitting, specifically including: S2231: Select a point cloud as a seed point from the complete power line data obtained after the fine extraction operation; S2232: Traverse all laser point clouds except the seed point and calculate the distance between each point cloud and the seed point. Find its nearest neighbor in terms of spatial distance and record its number. ; S2233: Set a distance threshold If the distance is obtained according to S2232 Less than Then the number is Place the point and the seed point together; then The point is used as the new seed point, and the remaining points are traversed. S2234: When the distances between the remaining points and the seed point do not meet the distance threshold. This indicates that the identification of a complete power line point cloud has been completed. S2235: After identifying a power line, randomly select one point from the remaining points as a new seed point, and repeat S2232, S2233 and S2234 until no new point can be found as a seed point; complete the identification and separation of each power line in the complete power line system.

7. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 6, characterized in that: S221 spatial feature constraint filtering specifically includes: Transform the point cloud of ground features into a 3D grid and then convert the voxels containing the points into a grid. Marked as 1; All voxels with the same position on the plane are considered as the basic unit of judgment, i.e. directional cubic pillars, and the size of the interval of statistical blank voxels based on the presence or absence of data points; For a single column If there is only one layer of point cloud, then this cubic column does not contain power line point cloud; if there are multiple layers of points and the interlayer interval is less than the interval threshold finterval, then this column does not contain power line points; if there are multiple layers of points and the interlayer interval is greater than the interval threshold finterval, then the points below the layer with the largest interlayer interval are removed, and the remaining points are candidate power line points.

8. The method for spatial distance detection of power transmission lines based on lidar and vision measurement according to claim 3, characterized in that: S23 specifically refers to the following: further optimizing the point cloud extraction algorithm for transmission lines, adopting a feature fusion-based method to fuse features in the image and point cloud data; achieving accurate extraction, strand division, and fitting of transmission lines, and finally obtaining a complete three-dimensional model of the transmission line.

9. An electronically readable storage medium having a computer program / instructions stored thereon, characterized in that, The computer program / instructions are executed by the processor to perform the spatial distance detection method for power transmission lines based on lidar and visual measurement as described in any one of claims 1-8.

10. An electronic device, characterized in that, The electronic device includes: processor; A memory having a computer program stored thereon, the computer program being executed by the processor to perform the steps of the method for detecting spatial distances of power transmission lines based on lidar and visual measurement as described in any one of claims 1-8.

Images