Insulator Fault Detection in Aerial Images Using YOLOv9

Insulator faults are a severe threat to the safety and stability of power transmission lines. Accurate and fast detection of these faults is paramount for preventive maintenance and grid reliability. This paper explores the application of You Only Look Once version 9 (YOLOv9), a state-of-the-art deep learning object detection model, for insulation fault detection in power line imagery. The effectiveness of YOLOv9 in identifying healthy insulators from those exhibiting various fault types is investigated. The paper details the training process, including data preparation, model configuration, and optimization techniques. Standard metrics like mean average precision (mAP) are used to analyze the performance of the YOLOv9 mode and the confusion matrix is used to identify potential areas for improvement. The findings of this research contribute to the development of automated and efficient insulation fault detection systems, enhancing power grid maintenance strategies.