Electric-power line detection method and system based on millimeter wave radar decision tree classification

Abstract

The invention discloses an electric-power line detection method and system based on millimeter wave radar decision tree classification. The electric-power line detection method based on millimeter wave radar decision tree classification includes the steps: utilizing a millimeter wave radar carried on an unmanned plane to acquire a radar image of an electric-power line; transformation of coordinates: utilizing a method of transformation of coordinates to transform the radar image of the electric-power line to a rectangular plane coordinate system from a polar coordinate system; image processing by block: taking the distance in the image as the standard, setting a threshold, and blocking the image according to the set threshold; utilizing Hough transformation to detect the linear segments existing in the image blocks; obtaining a candidate linear segment set; and utilizing a decision tree classification model to classify the linear segments in the candidate linear segment set so as to detect the position of the electric-power line. The electric-power line detection method and system based on millimeter wave radar decision tree classification have the advantages of being able to effectively detect the electric-power line target in real time and being relatively low in the false alarm probability.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a power line detection method and system based on millimeter-wave radar decision tree classification. Background technique [0002] Ultra-high voltage transmission lines have always been responsible for the important task of power transmission in my country. The inspection and management of transmission lines is a basic work to effectively ensure the safety of transmission and distribution lines and their equipment. In order to meet the needs of daily inspection, UAV inspection is introduced into the inspection task as a routine inspection mode. However, affected by many factors such as terrain and weather conditions, the image data of the UAV may not be transmitted back to the ground station, causing the ground operator to lose effective control of the UAV, thus causing the danger of the UAV hitting the line. And then cause a safety accident. In order to ensure the in...

AI Technical Summary

Problems solved by technology

However, LiDAR equipment is usually large in size and weight, and drones usually have limited loads. It is impossible to assemble LiDAR on ordinary drones for data collection, and the collected LiDAR needs to be analyzed by special software, which takes a long time to process. Unable to meet the requirements of real-time identification and detection of power lines during the inspection process

[0006] Chinese invention patent CN 101625723A realizes the real-time extraction technology of power lines based on visible light images, but because the transmission lines are in the field with complex backgrounds, and are affected by factors such as illumination and shooting angles, the power lines are not obvious in the image, resulting in a decline in detection efficiency

Images