The Application of an Improved YOLOv8 Model Based on Deep Information Fusion in Power Transmission Line Inspection

With the increasing complexity of the transmission line environment, how to efficiently and accurately detect insulator defects and foreign objects on the transmission line has become the key to ensure the stability and safety of the power system. To this end, this paper proposes a power transmission line defect detection model based on YOLOv8, called PS-YOLO. Due to the significant difference in the area ratio between defects and the background in power transmission lines, existing detection methods perform poorly in identifying small target defects. To address this issue, this paper introduces a monocular depth estimation model, DPT (Dense Prediction Transformers), combined with the YOLOv8 object detection model, which enhances the model's ability to distinguish between targets and backgrounds. Additionally, this paper incorporates the ADown module and a multi-branch feature extraction network into YOLOv8, further improving its detection capability for targets of different scales. Experimental results show that PS-YOLO outperforms current mainstream models in detection accuracy and computational efficiency, providing an effective solution for power transmission line defect detection.