Foreign object detection of transmission line based on improved Yolov5

Abstract

The environment of the transmission line is complex and easy to attach foreign matter, which has always been one of the reasons for the safety hazards to the operation of the transmission line. In view of the current problem that the detection accuracy of the foreign matter on the transmission line needs to be improved, an improved transmission line foreign matter detection method based on the Yolov5 algorithm is proposed. Detection model. First, the RepVGG module is introduced into the feature extraction network to enhance the network feature extraction ability and improve the model reasoning speed; secondly, the ability of the network to identify important feature information is strengthened by integrating the attention mechanism module; finally, by adding the prediction layer and Soft-nms the algorithm processes the target prediction frame to improve the detection accuracy of the model. The experimental results show that the improved Yolov5 transmission line foreign object detection algorithm proposed in this paper has a mAP value of 4.1% higher than the conventional Yolov5, and it also has certain advantages in performance compared with the conventional target detection algorithm.