Abstract:The detection of transmission line equipment is of great significance for the safe and stable operation of the power system. With the development of social economy and the expansion of the scale of the power system, higher requirements are put forward for the safety and reliability of transmission lines. This paper proposes a method based on the improved YOLOv8 algorithm combined with the Pyramid Split Attention (PSA) mechanism to improve the detection accuracy of small targets of transmission line equipment. By introducing the PSA module, the model has been significantly improved in terms of multi-scale feature fusion and feature enhancement, thus effectively improving the perception ability of small targets. Experimental results show that on the InsPLAD-det dataset, this method achieves a mean Average Precision (mAP50) of 90.6% and has good real-time detection performance. In addition, ablation experiments verify the effectiveness of the PSA module and its adaptability in models of different scales.