Detection of Pantograph Angle Deficiency Based on Optimized YOLOv8n Model

A optimized YOLOv8 detection algorithm, UIB-YOLOv8, is proposed to address the problems of high missed and false detection rates in pantograph fault detection, as well as difficulty in identifying pantograph angles in environments with insufficient lighting. A key point detection algorithm based on a high-resolution feature pyramid network is used to extract the position of the pantograph angle. The UIB module is introduced into the backbone network to enhance feature extraction in the image. The MQA attention mechanism is combined to reduce the number of model parameters and improve detection speed. Finally, a Focal Loss loss function is added before the result output to reduce the problem of imbalanced sample categories. The experimental results show that UIB-YOLOv8 improves MAP by 2.2%, reduces parameter count by 26.35%, and reduces floating-point computation by 24.36% compared to YOLOv8 algorithm. Especially in low light conditions, the accuracy of identifying pantograph angle is improved by 4.8%.