UG-Net: An Unsupervised-Guided Framework for Railway Foreign Object Detection
Source abstract
Abstract
Foreign object intrusion severely threatens railway safety. Existing methods struggle with open-set categories, high annotation costs, and poor label-efficient generalization. To address these issues, we propose UG-Net, an unsupervised-guided label-efficient detection framework. The core idea is a two-stage strategy: first, a masked autoencoder (MAE) learns “normality” priors from unlabeled data and generates a spatial attention mask via a deep feature difference strategy; then, this mask is fused as a fourth channel into a lightweight YOLOv8n detector. This approach effectively alleviates reliance on manual annotations. On a self-constructed railway dataset, UG-Net achieved 94.56% mAP@0.5 using only 200 labeled samples, significantly outperforming the YOLOv8n baseline (86.91%). The framework provides a label-efficient solution for industrial anomaly detection.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Anomaly Detection Techniques and Applications · Advanced Neural Network Applications · Maritime Navigation and Safety
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Railway Transportation, Shanghai Institute of Technology, Shanghai 201400, China
References (25)
Show source-provided references
1. Qu, J., Li, S., Li, Y., and Liu, L. (2023). Research on Railway Obstacle Detection Method Based on Developed Euclidean Clustering. Electronics, 12.10.3390/electronics12051175 ↗
2. Design of Near Infrared Reflective Effective Pigment for LiDAR Detectable Paint—Addendum10.1557/adv.2020.115 ↗
3. Paulsen, R., and Pedersen, K. (2015). Detecting Rails and Obstacles Using a Train-Mounted Thermal Camera. Image Analysis. SCIA 2015, Springer. Lecture Notes in Computer Science.
4. A Cost-Effective Ultrasonic Sensor-Based Driver-Assistance System for Congested Traffic Conditions10.1109/TITS.2009.2026671 ↗
5. Laser-Based Obstacle Detection at Railway Level Crossings10.1155/2016/1719230 ↗
6. Robust Real-Time Unusual Event Detection Using Multiple Fixed-Location Monitors10.1109/TPAMI.2007.70825 ↗
7. Mukojima, H., Deguchi, D., Kawanishi, Y., Ide, I., Murase, H., Ukai, M., Nagamine, N., and Nakasone, R. (2016, January 25–28). Moving Camera Background-Subtraction for Obstacle Detection on Railway Tracks. Proceedings of the 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA.10.1109/ICIP.2016.7533104 ↗
8. Frontal Obstacle Detection Using Background Subtraction and Frame Registration10.2219/rtriqr.58.4_298 ↗
9. Efficient railway tracks detection and turnouts recognition method using HOG features10.1007/s00521-012-0846-0 ↗
10. Schölkopf, B., Williamson, R.C., Smola, A.J., Shawe-Taylor, J., and Platt, J. (December, January 29). Support Vector Method for Novelty Detection. Proceedings of the Advances in Neural Information Processing Systems 12 (NIPS), Denver, CO, USA.
11. Automatic Surveillance of People and Objects on Railway Tracks
12. Research on foreign object intrusion detection in railway tracks based on MSL-YOLO10.1186/s44147-025-00708-7 ↗
13. Railway Traffic Object Detection Using Differential Feature Fusion Convolution Neural Network10.1109/TITS.2020.2969993 ↗
14. Wang, C.Y., Bochkovskiy, A., and Liao, H.Y.M. (2023, January 17–24). YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPR52729.2023.00721 ↗
15. Wang, C., He, W., Nie, Y., Guo, J., Liu, C., Wang, Y., and Han, K. (2023, January 10–16). Gold-YOLO: Efficient Object Detector via Gather-and-Distribute Mechanism. Proceedings of the 37th International Conference on Neural Information Processing Systems, New Orleans, LA, USA.
16. An Improved Lightweight YOLOv5 Algorithm for Detecting Railway Catenary Hanging String10.1109/ACCESS.2023.3322444 ↗
17. High-speed railway foreign object intrusion detection algorithm based on improved YOLOv7
18. Liu, Z., Li, Z., Mofor, R.N., and Ning, D. (2020, January 18–21). Unsupervised Anomaly Detection in Railway Catenary Condition Monitoring Using Autoencoders. Proceedings of the 2020 IEEE Industrial Electronics Society Annual Conference (IECON), Singapore.
19. He, K., Chen, X., Xie, S., Li, Y., Dollár, P., and Girshick, R. (2022, January 18–24). Masked Autoencoders Are Scalable Vision Learners. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA.10.1109/CVPR52688.2022.01553 ↗
20. Li, C.L., Sohn, K., Yoon, J., and Pfister, T. (2021, January 20–25). CutPaste: Self-Supervised Learning for Anomaly Detection and Localization. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.10.1109/CVPR46437.2021.00954 ↗
21. Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery
22. Lyu, Y., Han, Z., Zhong, J., Li, C., and Liu, Z. (2019, January 20–23). A GAN-Based Anomaly Detection Method for Isoelectric Line in High-Speed Railway. Proceedings of the 2019 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Auckland, New Zealand.10.1109/I2MTC.2019.8827162 ↗
23. Gu, A., and Dao, T. (2023). Mamba: Linear-Time Sequence Modeling with Selective State Spaces. arXiv.
24. Zhu, L., Liao, B., Zhang, Q., Wang, X., Liu, W., and Wang, X. (2024, January 21–27). Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model. Proceedings of the 41st International Conference on Machine Learning (ICML), Vienna, Austria.
25. Johnson, J., Alahi, A., and Fei-Fei, L. (2016, January 11–14). Perceptual losses for real-time style transfer and super-resolution. Proceedings of the European Conference on Computer Vision (ECCV), Amsterdam, The Netherlands.10.1007/978-3-319-46475-6_43 ↗