Article Details

Abstract

To address safety risks such as personnel intrusion and foreign-object intrusion along high-speed railway perimeters and ensure train operation safety, this study analyzes perimeter intrusion monitoring technologies for high-speed railways. The study first classifies perimeter intrusion monitoring scenarios and reviews the basic principles and application characteristics of AI-based video monitoring technology. On this basis, the study focuses on the working principles and implementation approaches of three multi-sensor fusion technologies—vibration fiber plus AI video, millimeter-wave radar plus AI video, and LiDAR plus AI video—and evaluates their monitoring performance. The results indicate that multi-sensor fusion technologies enrich the dimensions of perceptual information and establish multi-source cross-validation mechanisms, thereby effectively improving the reliability of early warnings, compensating for the limitations of single-sensor technologies, and demonstrating good prospects for engineering applications.

Keywords

high-speed railway; perimeter intrusion; monitoring technology; multi-sensor fusion; Target Detection

References

1. [1] Liu Qinghong, Guan Zebin, Zhao Ying, etc Overview of High speed Railway Operation Environment Safety Monitoring System [J]. China Railway, 2023 (4): 40-47
2. [2] Hou Shengjie, Liu Xiankai, Tang Kaiyi, etc Research on Environmental and Disaster Monitoring and Early Warning System Scheme for High speed Maglev Transportation [J]. High speed Railway Technology, 2022, 13 (1): 7-11, 38
3. [3] Wang Rui Research on Key Technologies for High speed Railway Disaster and Intrusion Monitoring [D]. Beijing: China Academy of Railway Sciences, 2022
4. [4] JIANG Z F，WANG H，LUO G L，et al. Pedestrian Intrusion Detection in Railway Station Based on Mirror Translation Attention and Feature Pooling Enhancement [J].IEEE Signal Processing Letters，2024，31：2730-2734.
5. [5] Zhou X F，Yu Z J，Ruan T，et al. Robust IR-VIS Image Registration with Different FOVs in Railway Intrusion Detection [J/OL].Measurement，2024，225：113928.https://doi.org/10.1016/j.measurement.2023.113928.
6. [6] Chen Guohua Research on Expressway Abnormal Event Monitoring System Based on AI Video Recognition Technology [J]. Modern Information Technology, 2022, 6 (24): 18-22
7. [7] Bai Mingming, Yuan Zhenjiang, Zhou Jialiang A high-speed railway perimeter protection system based on vibration fiber optic and video analysis technology [J]. Information Technology and Informatization, 2019 (2): 72-74
8. [8] Aktas M，Akgun T，Demircin M U，et al. Deep Learning Based Multi-Threat Classification for Phase-OTDR Fiber Optic Distributed Acoustic Sensing Applications [C]//SPIE Commercial + Scientific Sensing and Imaging，2017，10208：102080G.
9. [9] Yao Feng, Wang Junfeng, Liu Hongjiang, etc Railway Area Security System Based on Radar Video Collaboration Technology [J]. China Railway, 2022 (4): 114-119
10. [10] Liu Zichang, Wang Junfeng, Lv Dayong Research on the Application of High speed Railway Perimeter Security System Based on Microwave Array Technology [J]. Railway Communication Signal Engineering Technology, 2019, 16 (2): 37-41, 59
11. [11] Wang Junfeng, Sheng Dajun, Liu Zichang, etc Research on the Application of High speed Railway Perimeter Intrusion Detection System Based on Microwave Array Technology [J]. China Railway, 2019 (2): 41-46
12. [12] Sha Yulin Exploration of the Application of Phased Array Radar Perimeter Detection Technology in Railways [J]. Railway Communication Signal Engineering Technology, 2019, 16 (2): 46-49
13. [13] Fu Huijin, Guo Pengyue, Xu Chengwei, etc Research on high-speed railway perimeter intrusion monitoring technology scheme based on multi-sensor technology fusion [J]. Railway Transport and Economy, 2022, 44 (9): 122-129
14. [14] Zhang Zhengyu High speed rail perimeter security detection based on deep learning [D]. Beijing: Beijing University of Posts and Telecommunications, 2022
15. [15] Dou Rui Design and Implementation of Railway Obstacle Monitoring System Based on Lidar [D]. Beijing: China University of Petroleum (Beijing), 2021
16. [16] Wang Tong, Liao Haoyu Research on the Application of edge computing in High speed Railway Perimeter Intrusion Monitoring [J]. Railway Computer Application, 2024, 33 (5): 40-45

Funding

China Railway Group Corporation Science and Technology Research and Development Program Project (P2021T001)

About the authors

Chen Guopeng (1973-), male, Han nationality, from Botou, Hebei, is a senior engineer with a bachelor's degree. His research direction is engineering measurement and control and technical management; Song Xiaoqi (1972-), male, Han ethnicity, born in Xi'an, Shaanxi Province, is a senior engineer with a bachelor's degree. His research direction is engineering measurement and control and technical management; Yuan Qianfang (1987-), male, Han ethnicity, from Tonglu, Zhejiang, is a senior engineering engineer with a bachelor's degree. His research focuses on video and the Internet of Things.