Title: Railway track area perception method based on computer vision

Authors: Dehua Wei; Zhi Li; Haijun Li; Yuxing Jiang

Addresses: School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou, 730070, China; Key Laboratory of Railway Industry on Plateau Railway Transportation Intelligent Management and Control, Lanzhou, 730070, China ' School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou, 730070, China ' Wuwei Vocational College, Wuwei, 733000, China; School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou, 730070, China; Key Laboratory of Railway Industry on Plateau Railway Transportation Intelligent Management and Control, Lanzhou, 730070, China ' School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou, 730070, China; Key Laboratory of Railway Industry on Plateau Railway Transportation Intelligent Management and Control, Lanzhou, 730070, China

Abstract: The rapid growth of railways and increase in operating mileage have imposed strict safety requirements. Foreign body intrusion is a major safety threat. Accurate perception of foreign body intrusion to a railway track area is crucial for railway safety. To meet this demand, an intelligent perception method for the railway track area is studied in this paper. Firstly, the image is denoised and its edge features are extracted; the rail lines are divided by morphological transformation. Secondly, the sliding window technology is used to locate the high pixel density segment. Thirdly, a corresponding polynomial fits a smooth curve to the pixels. In line with safety regulations, the area divided by the extended fitting line is calculated to get the required area. Finally, the obtained area is mapped to the original image for the detection task. Effectiveness of this method is verified with experiments.

Keywords: track segmentation; edge detection; morphological processing; feature analysis; railway safety.

DOI: 10.1504/IJVSMT.2025.147910

International Journal of Vehicle Systems Modelling and Testing, 2025 Vol.19 No.3, pp.262 - 284

Received: 10 Oct 2024
Accepted: 27 Feb 2025
Published online: 07 Aug 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article