An automatic extraction method of transmission line components for UAV inspection images

Abstract

The invention discloses a method for automatically extracting power transmission line components used for drone inspection images, which reads inspection image data, performs scaling processing, separates color images from RGB channel data, makes histogram statistics, traverses images, and determines The edge information of the image; the image is divided into blocks, the characteristics of the image block are extracted, and the similar features are aggregated to obtain the feature block; combined with the inherent attributes of the transmission line components, the feature analysis is performed on the feature block to determine the value of each feature block Class attributes to label components on transmission lines. The invention screens the images obtained by patrol inspection, reduces the burden of manual screening, and has broad application prospects in the patrol inspection field.

Description

technical field [0001] The invention belongs to the technical field of transmission line equipment monitoring, relates to digital image processing and pattern recognition technology, and in particular to an automatic extraction method for transmission line components used in drone inspection images. Background technique [0002] With the rapid development of my country's national economy and the increasing scale of urban construction, the demand for electric energy has increased. In order to ensure reliable power supply for enterprises and residents, power companies are required to conduct regular inspection and maintenance of transmission lines, especially the auxiliary equipment of the lines, so as to ensure the stable power supply and safe operation of the entire power grid. Due to the vast territory of our country, the transmission mileage of transmission lines is large, and the geographical environment is complex (mountains, rivers, hills, etc.), the operation safety of...

AI Technical Summary

Problems solved by technology

Existing methods are for identification or image segmentation of a certain part of the transmission line. For example, the patent CN105023014 A uses the strategy of corner detection to realize the positioning of the tower. This method is suitable for the inspection image of the tower with a relatively simple background. When shooting When the background color is similar to that of the tower or the background is complex (such as buildings, etc.), it will interfere with the positioning and reduce the accuracy of positioning; the patent CN101630411 A realizes the calculation of the pixel entropy value of the inspection image by calculating the power transmission line and the background segmentation, but did not realize the automatic recognition function of specific components, and this patent mainly deals with the inspection images with the sky as the background, and is not suitable for processing the UAV inspection images with the earth as the background; the patent CN101625723 A utilizes the improved Hough transform extracts the current features in the inspection image, and combines the template fitting method to realize the positioning of the wire in the image

Due to the fact that there are many postures of the wires in practice, a large number of templates need to be manually designed in order to realize the positioning of the wires, and the workload is relatively large

Images