Foreign Objects Identification of Transmission Line Based on Improved YOLOv7

As the grid coverage rises, foreign objects invade more and frequently, causing failures to rise every year. To address this issue, paper proposes a deep learning-based transmission line unmanned inspection of recognition algorithm. The algorithm is based on YOLOv7 (You Only Look Once) algorithm, combining with hyperparameter optimization genetic (GA) space-to-depth (SPD) convolution complete object Unmanned Aerial Vehicle (UAV) images. proposed method can promptly determine locate these targets’ presence in aerial Finally, compares improved other YOLO series algorithms (Faster-rcnn, Centernet, target detection models). comparison results show that has highest Mean Average Precision (mAP) 92.2% Frames Per Second (FPS) 19 second only Centernet. Compared unimproved YOLOv7, average accuracy tower cranes increased by 11.9%, which most obvious improvement compared targets. Meanwhile, speeds up convergence model.