CN-115578654-B - Point cloud-based power transmission line pole type identification method and system

CN115578654BCN 115578654 BCN115578654 BCN 115578654BCN-115578654-B

Abstract

The invention provides a transmission line tower type identification method based on point clouds, which comprises the steps of manufacturing a tower type data set, converting a standard tower model into an image file with consistent height and width, establishing a standard tower character library, extracting transmission tower point clouds, removing miscellaneous points and wire point clouds contained in the unmanned aerial vehicle real point clouds, extracting electric power tower point clouds, calculating a tower symmetry axis plane, calculating the symmetry axis plane of the tower real point clouds by counting points in all angle directions, re-projecting the tower point clouds, projecting the tower real point clouds to the tower symmetry axis plane in a coordinate transformation mode, identifying the tower type, and judging the tower type by adopting a template matching method. The invention improves the automation level of the point cloud data processing of the power transmission line, solves the problem of great manpower and material resources consumption for visual discrimination, and provides a data basis for subsequent automatic modeling and deformation detection of the pole tower.

Inventors

AI MINGYAO
HUANG SHIMAN
HU QINGWU
LIU TIANCHENG
LI JIAYUAN
ZHAO PENGCHENG

Assignees

武汉大学

Dates

Publication Date: 20260508
Application Date: 20220930

Claims (6)

1. A method for identifying the type of a power transmission line pole tower based on point cloud is characterized by comprising the following steps, Step 1, manufacturing a tower type data set, which comprises the steps of converting a standard tower model into an image file with consistent height and width, and establishing a standard tower character library; Step 2, transmission tower point cloud extraction, which comprises removing miscellaneous points and wire point clouds contained in the unmanned aerial vehicle real-time point cloud, and extracting electric power tower point clouds; step 3, calculating the symmetry axis of the tower, which comprises calculating the symmetry axis of the actual point cloud of the tower by counting the number of points in each angle direction; the implementation comprises the sub-steps of, Step 3.1, projecting the segmented pole tower point cloud to an XOY plane to obtain a pole tower plane position center point Namely, a local elevation maximum point of the tower point cloud; Step 3.2, setting step length by taking the central point of the position of the pole tower as the origin Sum interval N is recorded as the number of intervals, and each angle direction is counted Counting up and the direction with the largest number As a symmetry axis, pulling up along the Z axis to be used as a symmetry axis surface; step 4, tower point cloud heavy projection, which comprises projecting tower real point cloud to a symmetrical shaft plane of the tower in a coordinate transformation mode; the implementation comprises the sub-steps of, Step 4.1, calculating the central point P of the pole tower Wherein Taking the point P as a new origin of three-dimensional space coordinates, and a coordinate conversion calculation formula is as follows: Wherein, the In order to transform the pre-center point coordinates, The coordinates of the center point after conversion; step 4.2, three-dimensional coordinate rotation, including rotating the space coordinate system translated in step 4.1 into a space coordinate system with P as an origin and a symmetry axis plane as an XOY plane by space coordinate rotation, and remapping the Y axis to an original Z+ axis position, wherein an equation of the symmetry axis plane under the original space coordinate system is that The coordinate conversion formula is: Wherein, the In order to transform the pre-center point coordinates, In order to convert the coordinates of the center point, Is the angle of the symmetry axis; And 4.3, mapping the three-dimensional space coordinates converted in the step 4.2 onto an XOY plane through projection, wherein a coordinate conversion formula is as follows: X, Y, Z is the plane coordinates after projection; step 5, identifying the type of the tower, which comprises the step of judging the type of the tower by adopting a template matching method; the implementation comprises the sub-steps of, Step 5.1, binarizing the pole tower point cloud image, dividing X, Y values of all point clouds in a bag into grids, setting the width of a projection grid, taking 1 for the projection grid if point cloud pixel points exist in the projection grid, otherwise taking 0 for the projection grid, generating a binary image file with the same size as a standard pole tower character library, and recording the file size as M rows and N columns; Step 5.2, calculating the correlation coefficient R of the binary image file S (m, n) of the actual measuring tower and the file T (m, n) in the standard tower character library in sequence, when the actual measuring tower is matched with the tower type in the standard tower character library, the correlation coefficient has the maximum value, the calculation formula of the correlation coefficient is as follows, Wherein, the Is a file in an ith standard tower character library, S is a binary image file of an actual measured tower, The values of the m-th row and the n-th column of the file in the ith standard tower character library are obtained, For actually measuring the values of the m-th row and the n-th column of the binary image file of the tower, And the correlation coefficient between the file in the ith standard tower character library and the binary image file of the actual measured tower.
2. The method for identifying the type of the power transmission tower based on the point cloud as claimed in claim 1, wherein the implementation manner of the step 2 comprises the following substeps, Step 2.1, acquiring a three-dimensional point cloud of a pole tower scene by adopting an unmanned plane laser radar or inclination measurement technology; Step 2.2, removing the miscellaneous point data to obtain force patrol data containing electric towers, line ground and surrounding vegetation; step 2.3, gridding the point clouds, which comprises projecting the point clouds into a horizontal XY coordinate system along the vertical direction, dividing grids, judging the grid position of each point cloud, and realizing the ordering of the point clouds; step 2.4, extracting grid areas of the transmission towers, wherein the step comprises the steps of calculating local elevation maximum values, local elevation minimum values and local elevation differences in each segmented grid, setting elevation difference threshold values to remove grid areas containing non-transmission tower point clouds such as ground, low vegetation and the like in combination with the characteristic of large elevation differences of the transmission towers, removing areas where the transmission lines are located according to the continuity characteristics of the transmission tower point clouds on elevation distribution; And 2.5, extracting the point cloud of the tower, namely taking the maximum point of the local elevation of the point cloud at the top part of the tower as the center, extracting the point cloud at all elevations in a slightly larger range than the grid around, calculating Gao Chengzui small values in the grid as ground elevations, setting as a threshold value, removing the ground points and the point cloud of the tower feet, and extracting the point cloud of the tower head and the tower body of the tower.
3. A power transmission tower type identification system based on point cloud is characterized by being used for realizing the power transmission tower type identification method based on point cloud according to any one of claims 1-2.
4. The point cloud based power transmission tower type identification system of claim 3, comprising a module, The first module is used for manufacturing a tower type data set and comprises the steps of converting a standard tower model into an image file with consistent height and width, and establishing a standard tower character library; the second module is used for extracting the transmission tower point cloud, and comprises the steps of removing miscellaneous points and wire point cloud contained in the unmanned aerial vehicle real-time point cloud and extracting the power tower point cloud; the third module is used for calculating the symmetrical axis of the tower, and calculating the symmetrical axis of the actual point cloud of the tower by counting the number of points in each angle direction; the fourth module is used for tower point cloud heavy projection and comprises the step of projecting the tower real point cloud to the symmetrical shaft surface of the tower in a coordinate transformation mode; And a fifth module for identifying the type of the tower, comprising judging the type of the tower by adopting a template matching method.
5. The system of claim 3, wherein the processor and the memory are configured to store program instructions, and the processor is configured to invoke the stored instructions in the memory to perform a method of identifying a type of a power transmission tower based on a point cloud as claimed in any one of claims 1-2.
6. The point cloud based power transmission tower type identification system according to claim 3, comprising a readable storage medium, wherein the readable storage medium has a computer program stored thereon, and the computer program when executed implements a point cloud based power transmission tower type identification method according to any one of claims 1-2.

Description

Point cloud-based power transmission line pole type identification method and system Technical Field The invention relates to the technical field of point cloud data processing, in particular to a method and a system for identifying types of power transmission towers based on point clouds. Background With the development of smart grid technology, people put forward higher requirements on the digital and intelligent degree of grid management. By utilizing remote sensing monitoring based on SAR satellites, the method has the advantages of wide monitoring range, large area and the like, but has higher requirements on the sampling period and the image resolution of remote sensing images, and has huge data acquisition cost, complex data processing and difficult comprehensive popularization. In recent years, the recovery of the three-dimensional model of the overhead transmission line by using the airborne laser scanning and oblique photogrammetry technology has become a line inspection mode which is widely popularized in the national network system, and the line inspection efficiency and accuracy are remarkably improved. The tower type is very important tower modeling data, is a precondition for extracting characteristic points of the tower and identifying the state of the tower, and is visually distinguished by professional staff in the traditional method. However, with the development of laser scanning and oblique photogrammetry technologies, workers often need to rapidly process the point cloud data of hundreds of GB towers, and the manual visual mode is adopted to judge the towers at the moment, so that the workload of the data processor is increased, and the development trend of the automation of the point cloud data processing is not met. Therefore, a method for automatically identifying the type of the transmission tower based on the point cloud is needed. Disclosure of Invention The invention provides a transmission tower type identification scheme based on point clouds, which is characterized in that a standard tower character set is manufactured, the transmission tower point clouds are obtained, the symmetrical axial planes of the transmission tower point clouds are calculated, the transmission tower point clouds are subjected to reprojection, and the re-projected transmission tower point clouds are matched with a standard tower data set, so that the judgment of the type of the transmission tower is realized. In order to achieve the above purpose, the invention provides a method for identifying the type of a power transmission tower based on point cloud, which comprises the following steps: Step 1, manufacturing a tower type data set, which comprises the steps of converting a standard tower model into an image file with consistent height and width, and establishing a standard tower character library; Step 2, transmission tower point cloud extraction, which comprises removing miscellaneous points and wire point clouds contained in the unmanned aerial vehicle real-time point cloud, and extracting electric power tower point clouds; step 3, calculating the symmetry axis of the tower, which comprises calculating the symmetry axis of the actual point cloud of the tower by counting the number of points in each angle direction; step 4, tower point cloud heavy projection, which comprises projecting tower real point cloud to a symmetrical shaft plane of the tower in a coordinate transformation mode; and 5, identifying the type of the tower, wherein the method comprises the step of judging the type of the tower by adopting a template matching method. Moreover, the implementation of step 2 comprises the sub-steps of, Step 2.1, acquiring a three-dimensional point cloud of a pole tower scene by adopting an unmanned plane laser radar or inclination measurement technology; Step 2.2, removing the miscellaneous point data to obtain force patrol data containing electric towers, line ground and surrounding vegetation; step 2.3, gridding the point clouds, which comprises projecting the point clouds into a horizontal XY coordinate system along the vertical direction, dividing grids, judging the grid position of each point cloud, and realizing the ordering of the point clouds; Step 2.4, extracting grid areas of the transmission towers, wherein the step comprises the steps of calculating local elevation maximum values, local elevation minimum values and local elevation differences in each segmented grid, setting elevation difference threshold values to remove grid areas containing non-transmission tower point clouds such as ground, low vegetation and the like in combination with the characteristic of large elevation differences of the transmission towers, removing areas where the transmission lines are located according to the continuity characteristics of the transmission tower point clouds on elevation distribution; And 2.5, extracting the point cloud of the tower, namely taking the maximum point of the local elevation of t