Application of Enhanced YOLOv8 in Multi-object Detection for Autonomous Inspection of Transmission Lines

Overhead transmission lines, which are directly related to the safe and stable operation of 14 the power grid, are important channels for the transmission of electrical energy. Accurate and efficient 15 detection of electrical components in transmission lines is crucial for maintaining power grid stability 16 and safety. Existing methods often encounter difficulties with issues such as small targets, complex 17 backgrounds, and varying scales. In this paper, we propose an enhanced version of YOLOv8 18 specifically designed for multi-object detection in transmission line inspections. Secondly, with the aim 19 of enhancing the accuracy and robustness of electrical component identification as well as defect 20 detection, the Faster-Block module, GhostNet, SPD-Conv, and the triplet attention module are 21 incorporated into the original YOLOv8. Finally, improved YOLOv8 and the mainstream deep learning 22 algorithms are trained and tested on the self-built dataset. The mean average precision (mAP) value of 23 improved YOLOv8 reaches 86.73%, which is nearly 36%, 14% and 10% higher than that of Faster 24 R-CNN, DETR, YOLOv8, respectively. In addition, the inference speed of the improved YOLOv8 25 reaches 38.1 frames per second (FPS), which is 4.4% faster than that of YOLOv8. Ablation studies and 26 comparative experiments demonstrate the effectiveness of our approach in detecting insulators, 27 dampers, clamps, and defects, especially for small targets and complex scenes.