Two-Stage Multi-Label Detection Method for Railway Fasteners Based on Type-Guided Expert Model
Source abstract
Abstract
Railway track fasteners, serving as critical connecting components, have a reliability that directly impacts railway operational safety. To address the performance bottlenecks of existing detection methods in handling complex scenarios with diverse fastener types and co-occurring multiple defects, this paper proposes a Type-Guided Expert Model-based Fastener Detection and Diagnosis framework (TGEM-FDD) based on You Only Look Once (YOLO) v8. This framework follows a “type-identification-first, defect-diagnosis-second” paradigm, decoupling the complex task: the first stage employs an enhanced YOLOv8s with Deepstar, SPPF-attention, and DySample (YOLOv8s-DSD) detector integrating Deepstar Block, Spatial Pyramid Pooling Fast with Attention (SPPF-Attention), and Dynamic Sample (DySample) modules for precise fastener localization and type identification; the second stage dynamically invokes a specialized multi-label classification “expert model” based on the identified type to achieve accurate diagnosis of multiple defects. This study constructs a multi-label fastener image dataset containing 4800 samples to support model training and validation. Experimental results demonstrate that the proposed YOLOv8s-DSD model achieves a remarkable 98.5% mean average precision at an Intersection over Union threshold of 0.5 (mAP@0.5) in the first-stage task, outperforming the original YOLOv8s baseline and several mainstream detection models. In end-to-end system performance evaluation, the TGEM-FDD framework attains a comprehensive Task mean average precision (Task mAP) of 88.1% and a macro-average F1 score for defect diagnosis of 86.5%, significantly surpassing unified single-model detection and multi-task separate-head methods. This effectively validates the superiority of the proposed approach in tackling fastener type diversity and defect multi-label complexity, offering a viable solution for fine-grained component management in complex industrial scenarios.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Railway Engineering and Dynamics · Railway Systems and Energy Efficiency · Infrastructure Maintenance and Monitoring
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mechanical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
Funding
National Natural Science Foundation of China · Lanzhou Science and Technology Plan Project
References (31)
Show source-provided references
1. Analysis of Elastic Bar Fracture Causes of Fasteners in Ballastless Track of High-Speed Railway
2. Review of Integrated Full Life Cycle Data Management and Application of the Slab Tracks
3. Bono, F.M., Radicioni, L., Cinquemani, S., Benedetti, L., Cazzulani, G., Somaschini, C., and Belloli, M. (2023). A Deep Learning Approach to Detect Failures in Bridges Based on the Coherence of Signals. Future Internet, 15.10.3390/fi15040119 ↗
4. Identification Method for Subgrade Settlement of Ballastless Track Based on Vehicle Vibration Signals and Machine Learning10.1016/j.conbuildmat.2023.130573 ↗
5. Wireless Rail Fastener Looseness Detection Based on MEMS Accelerometer and Vibration Entropy10.1109/JSEN.2019.2955378 ↗
6. Optimized Railway Track Health Monitoring System Based on Dynamic Differential Evolution Algorithm10.1016/j.measurement.2019.107332 ↗
7. A Real Time Detection Method of Track Fasteners Missing of Railway Based on Machine Vision
8. Learning Visual Similarity for Inspecting Defective Railway Fasteners10.1109/JSEN.2019.2911015 ↗
9. Deep Multitask Learning for Railway Track Inspection10.1109/TITS.2016.2568758 ↗
10. Wang, Z., and Wang, S. (2015, January 12–13). Research of Method for Detection of Rail Fastener Defects Based on Machine Vision. Proceedings of the 4th International Conference on Mechatronics, Materials, Chemistry and Computer Engineering 2015, Xi’an, China.10.2991/icmmcce-15.2015.547 ↗
11. Girshick, R. (2015, January 7–13). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.10.1109/ICCV.2015.169 ↗
12. Ren, S., He, K., Girshick, R., and Sun, J. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. Advances in Neural Information Processing Systems 28, Neural Information Processing Systems Foundation, Inc. (NeurIPS).
13. Railway Track Fastener Defect Detection Based on Image Processing and Deep Learning Techniques: A Comparative Study10.1016/j.engappai.2019.01.008 ↗
14. MYOLOv3-Tiny: A New Convolutional Neural Network Architecture for Real-Time Detection of Track Fasteners10.1016/j.compind.2020.103303 ↗
15. Real-Time Railroad Track Components Inspection Based on the Improved YOLOv4 Framework10.1016/j.autcon.2021.103596 ↗
16. Fu, J., Chen, X., and Lv, Z. (2022). Rail Fastener Status Detection Based on MobileNet-YOLOv4. Electronics, 11.10.3390/electronics11223677 ↗
17. Li, X., Wang, Q., Yang, X., Wang, K., and Zhang, H. (2023). Track Fastener Defect Detection Model Based on Improved YOLOv5s. Sensors, 23.10.3390/s23146457 ↗
18. A Rail Fastener Defect Detection Algorithm Based on Improved YOLOv510.1177/09544097241234380 ↗
19. Cascade Learning Embedded Vision Inspection of Rail Fastener by Using a Fault Detection IoT Vehicle
20. Convolutional Neural Network for Detecting Railway Fastener Defects Using a Developed 3D Laser System10.1080/23248378.2020.1825128 ↗
21. Autonomous Structural Visual Inspection Using Region-Based Deep Learning for Detecting Multiple Damage Types10.1111/mice.12334 ↗
22. DenseSPH-YOLOv5: An Automated Damage Detection Model Based on DenseNet and Swin-Transformer Prediction Head-Enabled YOLOv5 with Attention Mechanism10.1016/j.aei.2023.102007 ↗
23. Varghese, R., and Sambath, M. (2024, January 4–5). YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness. Proceedings of the 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), Chennai, India.10.1109/ADICS58448.2024.10533619 ↗
24. Terven, J., and Cordova-Esparza, D. (2023). A Comprehensive Review of YOLO: From YOLOv1 and Beyond. Electronics, 12.
25. Wang, C.-Y., Bochkovskiy, A., and Liao, H.-Y.M. (2022, January 18–24). 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), New Orleans, LA, USA.10.1109/CVPR52729.2023.00721 ↗
26. Doublem-Net: Multi-Scale Spatial Pyramid Pooling-Fast and Multi-Path Adaptive Feature Pyramid Network for UAV Detection10.1007/s13042-024-02278-1 ↗
27. Li, H., Li, J., Wei, H., Liu, Z., Zhan, Z., and Ren, Q. (2022). Slim-Neck by GSConv: A Better Design Paradigm of Detector Architectures for Autonomous Vehicles. Sensors, 22.
28. Liu, W.Z., Lu, H., Fu, H.T., Xie, Y., Zhang, D., and Zhang, Y. (2023, January 1–6). Learning to Upsample by Learning to Sample. Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France.10.1109/ICCV51070.2023.00554 ↗
29. Wang, J., Chen, K., Xu, R., Liu, Z., Loy, C.C., and Lin, D. (November, January 27). CARAFE: Content-Aware ReAssembly of FEatures. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.
30. Shi, W., Caballero, J., Huszár, F., Totz, J., Aitken, A.P., Bishop, R., Rueckert, D., and Wang, Z. (2017, January 21–26). Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2016.207 ↗
31. Zhu, X., Hu, H., Lin, S., and Dai, J. (2019, January 16–20). Deformable ConvNets V2: More Deformable, Better Results. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00953 ↗