A deep learning-based UAV intelligent inspection cable surface defect detection method

Abstract

The present invention relates to a deep learning-based UAV intelligent inspection cable surface defect detection method, which uses the YOLOv3 surface defect detection model to perform training and detection respectively, and uses the UAV to collect and verify cable image data and apply the constructed defect detection model , Evaluate and analyze application errors, construct a reasonable and effective evaluation mechanism, the invention can improve the identification accuracy of defects in cables, have a better ability to identify defects in cables at the initial stage, and can detect cable defects in time and quickly deal with them.

Description

Technical field [0001] The present invention relates to the field of image processing technology, particularly to a deep learning-based UAV intelligent inspection cable surface defect detection method. Background [0002] Today, the scale of China's power grid has jumped to the first place in the world. At present, six inter-provincial power grids have been built, namely the six major power grids in the south, northwest, east China, central China, north China and northeast China, with a total length of transmission lines exceeding 1.15 million kilometers, and transmission lines of 500kV and above have become the main transmission force of power grids in various regions. Due to the complex terrain, many hills, few plains, and complex and changeable meteorological conditions, the power line corridor needs to cross a variety of complex geographical environments, especially for the power line through the edge of the original forest, high altitude, ice and snow cover areas, and there ...

AI Technical Summary

Problems solved by technology

[0007] S1: The present invention uses a defect detection model to detect defects in cables, establishes a cable defect image database, marks all the images of cables, and then divides the training set and test set according to the preset ratio, and at the same time aims at the fact that a large number of actual cables cannot be obtained in practice Defect images are artificially synthesized to expand the data set;

Images