Development of a System for Detecting Defects in Power Transmission Poles

Abstract

This article offers a solution to the problem of diagnosing power transmission poles using the example of detecting defects in insulators. The main goal is to reduce the influence of the human factor on the diagnostic process, increase the accuracy of defect recognition and reduce the time for inspection of power lines. The article defines the problem, describes the methods and algorithms used. The process of working with the machine learning model, the data preparation process, and the main set of metrics for evaluating the quality of defect recognition are described sequentially. At the end of the paper, the results of the study are presented, including an analysis of the recognition quality using a basic set of metrics and visual images. The proposed algorithm for the operation of the aircraft is also described.