Research on the Application of YOLO v3 in Railway Intruding Objects Recognition
Yongtian Ma, Jianjun Fang, Jiaxiang Zhao, Qiushi Zhang · 2022 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA) · 2022
In order to detect foreign objects intruding into the track and prevent foreign objects from causing railroad safety accidents, the track foreign object intrusion detection algorithm is investigated. For the specific application scenario of railway foreign object intrusion, the enhanced YOLOv3 high speed railway foreign object detection network is proposed to improve the ability of using picture features and detection effect, and the average detection accuracy reaches 79.2% with slightly reduced detection speed, which is 4.4% higher than the original network. The enhanced YOLOv3 railroad foreign object intrusion detection network can effectively improve the detection accuracy of targets at different scales.