Research on Transmission Line Fault Hidden Danger Detection Model Based on Improved YOLOv5n
In order to realize accurate and efficient identification of transmission line fault hazards,this paper proposes a transmission line fault detection model based on improved YOLOv5n.First,the backbone network is replaced with FasterNet network,which not only reduces redundant computation by local convolutional PConv,but also effectively extracts features to enhance the model's ability to express important features.Secondly,combining the C2f module and Res2net structure using the ELAN idea,the C2f-Res2 module is proposed to perform multi-scale feature fusion on the input image at a deeper granularity level to further enhance the feature extraction capability of the network.Finally,the original loss function CIoU is replaced using the WIoU loss function to improve the quality of the anchor frame.The experimental results show that the average accuracy of this improved YOLOv5n transmission line fault detection model is 93.2%and the detection speed is 94.2 frames-s-1.It effectively reduces the errors and omissions in the detection of transmission line faults,and can identify and localize the transmission line fault hazards more efficiently.
object detectionimage processingtransmission line fault detectionYOLOv5
万方数据
维普
