Autonomous Inspection Method of Transmission Line Based on Key Points and Machine Vision

The application of UAV and other technologies reduces the manual input of inspection work, but some inspection methods using template matching can not meet the industrial requirements of identification accuracy and efficiency. In order to improve the above defects, a transmission line autonomous inspection method based on key points and machine vision is studied. After processing the line image with machine vision technology, random Hough transform and key point algorithm are used to locate the abnormal possible segment. The abnormal mapping feature in CenterNet is improved to realize the abnormal inspection of transmission lines. When the inspection method is tested on the experimental line, the detection accuracy of the method is higher than that of the existing methods, and the detection time is shortened about 38.35%, and the inspection quality is improved.