Foreign Object Detection Method in Transmission Lines Based on Improved YOLOv8n

The smooth operation of power transmission lines is crucial for maintaining the stability of the power system. Foreign objects, such as bird nests, trash, and balloons, when attached to transmission lines, can cause short circuits or fires, resulting in severe operational obstacles for the power grid. Against this background, the development of an efficient and accurate detection method is of great value in ensuring the safe operation of transmission lines. This paper presents a new method for detecting foreign objects on transmission lines using an improved version of the YOLOv8n model. The method integrates the MSDA attention mechanism into the architecture of YOLOv8n, optimizing the feature fusion process, and enabling the model to effectively capture feature information at different scales. Additionally, by introducing the Focal-EIOU loss function to replace the original CIOU loss function, this method significantly improves the recognition accuracy of foreign objects. The results of testing on a real-world dataset of foreign objects on power transmission lines demonstrate a significant improvement in the accuracy of object detection using this method. Specifically, the mAP0.5 metric for trash detection increased from 89.9% to 99.0%, a gain of 9.1%. The overall mAP0.5 reached 97.7%, representing a 1.7% improvement compared to the original YOLOv8 model.