Abstract
In order to enhance the real-time perception of the environment of the train and ensure safety, it is necessary to detect intrusion. Most of the intrusion detection algorithms have large parameters and slow speed, so they cannot be well applied in high-speed trains. Therefore, this paper proposes a lightweight pedestrian intrusion detection algorithm with on-board video. First, the lightweight object detection algorithm is used to realize pedestrian detection in the whole scene. Second, the railway track area is extracted by a lightweight semantic segmentation algorithm. Finally, by judging the coordinate position, the pedestrian intrusion detection in the railway track area is realized. In experiments, the positive detection rate is 94.84%, and the missing detection rate is 0.97%, and the false detection rate is 0.81%, which can effectively detect the pedestrian intrusion in the railway track and further improve the intelligent perception system of railway operation environment, which has practical significance and application value for railway safety.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Qin, Y., Cao, Z., Sun, Y.: Research on active safety methodologies for intelligent railway systems. Engineering (2022)
Li, X.: Research on Algorithm of railway foreign body Detection based on transfer learning. Beijing Jiaotong University (2016). (in Chinese)
Cao, Z., Qin, Y.: An effective railway intrusion detection method using dynamic intrusion region and lightweight neural network. Measurement 191, 110564 (2022)
Wang, Y., Yu, Z., Zhu, L.: A fast feature extraction algorithm for foreign bodies invaded by high-speed railway based on CNN. Chin. J. Sci. Instrum. 38(5), 1267–1275 (2017). (in Chinese)
Cao, Z., Qin, Y., Jia, L.: Haze removal of railway monitoring images using multi-scale residual network. IEEE Trans. Intell. Transp. Syst. 1–14 (2020)
Zhang, C.: Research on Algorithm of orbital Foreign Body Intrusion Detecton based on Deep learning. Jiangxi University of Science and Technology (2019). (in Chinese)
Bochkovskiy, A., Wang, C., Liao, H.: YOLOv4: Optimal speed and accuracy of object detection. In: IEEE Conference on Computer Vision and Pattern Recognition. IEEE, Washington, p. 10934 (2020)
Chen, L., Zhu, Y., Papandreou, G.: Encoder-decoder with atrous separable convolution for semantic image segmentation. In: Proceedings of the European conference on computer vision (ECCV), pp. 801–818. IEEE, Munich (2018)
Joseph, R., Ali, F.: YOLOv3: an incremental improvement. In: IEEE Conference on Computer Vision and Pattern Recognition, pp 1–6. IEEE, Salt Lake City (2018)
Liu, W., Anguelov D., Erhand D.: SSD: single shot multibox detector. In: Proceedings of the European Conference on Computer Vision (CVPR). Las Vegas (2016)
Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp 91–99. (2015)
Yu, C., Wang, J., Peng, C., Gao, C., Yu, G., Sang, N.: BiSeNet: bilateral segmentation network for real-time semantic segmentation. In: Proceedings of the European conference on computer vision (CVPR), pp. 325–341. (2018)
Poudel, R., Liwicki, S., Cipolla, R.: Fast-scnn: fast semantic segmentation network arXiv:1902.04502 (2019)
Acknowledgments
This work is supported by the National Key R&D Program of China (Contract No. 2022YFB4300601).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 Beijing Paike Culture Commu. Co., Ltd.
About this paper
Cite this paper
Gao, Y. et al. (2024). A Lightweight Pedestrian Intrusion Detection Algorithm Based on On-Board Video. In: Gong, M., Jia, L., Qin, Y., Yang, J., Liu, Z., An, M. (eds) Proceedings of the 6th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2023. EITRT 2023. Lecture Notes in Electrical Engineering, vol 1138. Springer, Singapore. https://doi.org/10.1007/978-981-99-9319-2_8
Download citation
DOI: https://doi.org/10.1007/978-981-99-9319-2_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-9318-5
Online ISBN: 978-981-99-9319-2
eBook Packages: EngineeringEngineering (R0)