The railway environment changes frequently and trains run fast. The railway-assisted driving system plays an essential role in railway safety. Pedestrians are prone to intrusion when the train enters and exits the platform. Therefore, we propose an algorithm for real-time detection of pedestrian intrusion in the track using onboard video. First, we use the yolov5 object detection algorithm to detect whether there is a pedestrian intrusion. If there is an intrusion, the Fast-SCNN algorithm is called to segment the railway track area. The compound algorithm judges whether the detected object is in the railway track area and reports an abnormality if there is a pedestrian intrusion. The detection speed of the compound detection algorithm proposed in this paper is 34-90FPS (frames per second), and the accuracy rate is 98.17%, which meets the requirements of real-time detection of track perimeter safety and assists driving in stations.
Railway Pedestrian Intrusion Detection Using Onboard Forward-Viewing Camera
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Chapter : 44 ; 388-396
2022-02-23
9 pages
Article/Chapter (Book)
Electronic Resource
English
Railway Pedestrian Intrusion Detection Using Onboard Forward-Viewing Camera
British Library Conference Proceedings | 2022
|