Railway intrusion seriously threatens railway safety and may cause serious casualties and huge property losses. With the development of autonomous driving, image processing algorithms based on vehicle front video have developed rapidly. This paper proposes a railway intrusion detection framework based on vehicle front video. The framework mainly consists of two stages: object detection and semantic segmentation. In the first stage, an object detection algorithm is used to detect potential intrusion objects. The second stage uses the semantic segmentation algorithm to obtain the railway perimeter area. If the object is found within the railway perimeter, it is regarded as an intrusion. The railway intrusion detection framework proposed in this paper starts the semantic segmentation algorithm only when the object detection algorithm detects potential intrusion objects. Finally, the proposed framework is tested on railway video and achieves 100% accuracy and 23 FPS (Frames Per Second).
A Railway Intrusion Detection Framework Based on Vehicle Front Video
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 : 5 ; 38-44
2022-02-19
7 pages
Article/Chapter (Book)
Electronic Resource
English
A Railway Intrusion Detection Framework Based on Vehicle Front Video
British Library Conference Proceedings | 2022
|Railway vehicle chassis front-end module, railway vehicle chassis and railway vehicle
European Patent Office | 2024
|