Detection of Suspended Solids in Transmission Lines Based on Improved YOLOv7s

With the UAV(Unmanned Aerial Vehicle) inspection gradually replacing manual inspection, how to accurately identify suspended solids from the complex background image of transmission lines is a big problem for UAV inspection. This work refines the precision of suspended solids detection within transmission lines by advancing the YOLOv7 framework, focusing on two key areas: data augmentation and the incorporation of an attentive mechanism. The data enhancement method in this paper is mainly divided into two parts, the first part is to simulate the random jitter of UAV aerial photography, and the second part is to simulate the noise addition and brightness change of shooting scene, environment, weather and other changes. Combined with the SimAM self-attention mechanism, the accuracy of the model is further improved, and the final mAP reaches 93.4%, which is 2.3% higher than the original model.