Research on identification and detection of foreign body intrusion for overhead lines

In this paper, a method of automatic identification and detection of foreign body intrusion for overhead transmission lines is proposed, which can solve the problem of heavy workload in daily inspection for overhead transmission lines. The actual overhead line images captured by inspection drones were taken as the data set, Label Image software was used to label the data set, and YOLOv5 model was used to train the data set. The test results show that the method proposed in this paper can accurately identify typical foreign bodies (such as bird nests, branches and textiles) in overhead lines, and accurately mark the location of foreign bodies at different shooting distances. The training accuracy of the algorithm reaches 97.7%, and the detection accuracy is maintained at a high level, which can meet the requirements of automatic operation and maintenance inspection for overhead transmission lines.