Transmission Line External Failure Non-contact Measurement Technology Based on YOLOv8-TP and Monocular Imaging

To address the challenges of accurately detecting external fault targets on power transmission lines and the issue of non-contact measurement, a non-contact measurement technology based on YOLOv8-TP and monocular imaging is proposed. First, a collaborative attention mechanism is introduced into the YOLOv8 backbone network to capture cross-channel information, directional awareness, and positional awareness. Next, a small-target detection layer is added to the original three-layer detection network to enhance the detection performance for small-sized potential external fault targets. Finally, a fusion of lower-level and higher-level feature maps is introduced in the PAN (Path Aggregation Network), ensuring that the model maintains strong semantic information while preserving detailed perception capabilities. Additionally, monocular imaging-based distance measurement technology is used to calculate the distance between the tower, conductor targets, and potential external fault factors based on key points. This method effectively improves the intelligent operation and maintenance level of the power system, playing a crucial role in enhancing the safety factor of transmission lines and reducing the need for labor.