Comparative Study on Rail Damage Recognition Methods Based on Machine Vision
Abstract
1. Introduction
2. On-Site Acquisition of Rail Damage
2.1. Acquisition Equipment for Rail Damage
2.2. Rail Damage Dataset
3. Deep Learning Networks for Rail Damage Detection
3.1. Object Detection Models
3.1.1. YOLO Model
3.1.2. SSD Model
3.1.3. Faster R-CNN Model
3.2. Detecting Performance Evaluation Indicators
4. Comparative Study on Rail Damage Recognition Methods
4.1. Comparative Analysis
4.2. Optimization of YOLO
4.2.1. Optimization Method
4.2.2. Optimization Results
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Zhao, L.; Mou, M.; Chen, D.; Zhong, M. Research on the influence of surface defects under the influence of rail corrosion on the fatigue damage of wheel rolling contact. Coatings 2025, 15, 589. [Google Scholar] [CrossRef]
- Cui, X.; Yin, Y.; Xu, X.; Peng, S.; Qi, Y.; Linghu, J.; Qi, W. Comparative Study on the Causes of Rail Corrugations in Long Steep Grade Sections Under Traction and Braking Conditions. Veh. Syst. Dyn. 2025, 63, 964–977. [Google Scholar] [CrossRef]
- Li, T.; Cui, X.; Zhao, X.; Zhang, H.; Li, J.; Qi, W. Three-Dimensional Reconstruction Method of Rail Corrugation Based on Vision Detection. Tribol. Trans. 2024, 67, 411–422. [Google Scholar] [CrossRef]
- Sung, D.; Hong, S.; Lee, J. Numerical analysis of the rail surface crack propagation under rail uplift force: A parametric study on initial crack geometry. Eng. Fail. Anal. 2023, 153, 107542. [Google Scholar] [CrossRef]
- Cui, X.; Zhong, Y.; Ding, H.; Zhang, H.; Li, X.; Linghu, J.; Guo, L. Evolution Mechanism of Rail Corrugation in Small-Radius Curve Section of Mountainous City Metro. Wear 2025, 570, 205943. [Google Scholar] [CrossRef]
- Lu, C.; Yin, J.; Zhang, Q.; Fu, Q. Wear analysis of brake lining in running-in stage for rail vehicles. Surf. Technol. 2023, 51, 63–71. [Google Scholar] [CrossRef]
- Pham, D.; Ha, M.; Xiao, C. A Novel Visual Inspection System for Rail Surface Spalling Detection. In Proceedings of the 2020 7th International Conference on Advanced Materials, Mechanics and Structural Engineering, Taipei, Taiwan, 25–27 September 2020. [Google Scholar] [CrossRef]
- Wu, F.; Wei, Y.; Li, Q. Damage detection and parameter learning method for high speed rail ultrasonic imaging. Comput. Integr. Manuf. Syst. 2021, 27, 747–756. [Google Scholar]
- Campos-Castellanos, C.; Gharaibeh, Y.; Mudge, P.; Kappatos, V. The application of long range ultrasonic testing (LRUT) for examination of hard to access areas on railway rails. In Proceedings of the 5th IET Conference on Railway Condition Monitoring and Non-Destructive Testing, London, UK, 28–30 November 2011. [Google Scholar] [CrossRef]
- Rizzo, P.; Cammarata, M.; Bartoli, I.; Di Scalea, F.L.; Salamone, S.; Coccia, S.; Phillips, R. Ultrasonic guided waves-based monitoring of rail head: Laboratory and field tests. Adv. Civ. Eng. 2010, 2010, 291293. [Google Scholar] [CrossRef]
- Liu, W.; Wang, S.; Yin, Z.; Tang, Z. Structural Damage Detection of Switch Rails Using Deep Learning. NDT E Int. 2024, 147, 103205. [Google Scholar] [CrossRef]
- Dong, L. Research on Nondestructive Detection of Electromagnetic Fusion Based on BP Neural Network. Ph.D. Thesis, Chongqing University, Chongqing, China, 2018. [Google Scholar]
- Jia, Y.; Liang, K.; Wang, P.; Ji, K.; Xu, P. Enhancement method of magnetic flux leakage signals for rail rail surface defect detection. IET Sci. Meas. Technol. 2020, 14, 711–717. [Google Scholar] [CrossRef]
- Lonkwic, P.; Krakowski, T.; Ruta, H. Use of Magnetic Flux Leakage to Diagnose Damage to a Lift Guide Rails System with Reference to the Sustainability Aspect. Sustainability 2024, 16, 1980. [Google Scholar] [CrossRef]
- Gong, W.; Akbar, M.F.; Jawad, G.N.; Shrifan, N.; Zhang, F. A magnetic field concentration method for magnetic flux leakage detection of rail-top surface cracks. IEEE Access 2024, 12, 43245–43254. [Google Scholar] [CrossRef]
- Nafiah, F.; Sophian, A.; Khan, M.R.; Abidin, I.M.Z. Quantitative evaluation of crack depths and angles for pulsed eddy current non-destructive testing. Non-Destr. Test. Eval. Int. 2019, 102, 180–188. [Google Scholar] [CrossRef]
- Heckel, T.; Thomas, H.-M.; Kreutzbruck, M.; Rühe, S. High speed non-destructive rail testing with advanced ultrasound and eddy-current testing techniques. In Proceedings of the NDTIP Proceedings, Prague, Czech Republic, 12–14 October 2009. [Google Scholar]
- Alvarenga, T.; Carvalho, A.; Honorio, L.; Cerqueira, A.; Filho, L.; Nobrega, R.A. Detection and Classification System for Rail Surface Defects Based on Eddy Current. Sensors 2021, 21, 7937. [Google Scholar] [CrossRef]
- Song, Z.; Yamada, T.; Shitara, H.; Takemura, Y. Detection of damage and crack in railhead by using eddy current testing. J. Electromagn. Anal. Appl. 2011, 3, 546–550. [Google Scholar] [CrossRef]
- Wu, F.; Mao, Z.C. Image enhancement and segmentation algorithm for rail surface defects. Comput. Simul. 2015, 32, 159–162. [Google Scholar] [CrossRef]
- Li, Q.; Ren, S. A real-time visual inspection system for discrete surface defects of rail heads. IEEE Trans. Instrum. Meas. 2012, 61, 2189–2199. [Google Scholar] [CrossRef]
- Hsieh, C.-C.; Hsu, T.-Y.; Huang, W.-H. An online rail rail fastener classification system based on YOLO models. Sensers 2022, 22, 9970. [Google Scholar] [CrossRef]
- Feng, J.H.; Yuan, H.; Hu, Y.Q.; Lin, J.; Liu, S.W.; Luo, X. Research on deep learning method for rail surface defect detection. IET Electr. Syst. Transp. 2020, 10, 436–442. [Google Scholar] [CrossRef]
- Guan, L.; Jia, L.; Xie, Z.; Yin, C. A lightweight framework for obstacle detection in the railway image based on fast region proposal and improved YOLO-tiny network. IEEE Trans. Instrum. Meas. 2022, 71, 5009116. [Google Scholar] [CrossRef]
- Kaewunruen, S.; Adesope, A.; Huang, J.; You, R.; Li, D. AI-Based Technology to Prognose and Diagnose Complex Crack Characteristics of Railway Concrete Sleepers. Discov. Appl. Sci. 2024, 6, 217. [Google Scholar] [CrossRef]
- Foret, P.; Kleiner, A.; Mobahi, H.; Neyshabur, B. Sharpness-aware minimization for efficiently improving generalization. arXiv 2020, arXiv:2010.01412. [Google Scholar] [CrossRef]
- Ding, L.; Rao, Z.Q.; Ding, B.; Li, S.J. Research on defect detection method of railway transmission line insulators based on GC-YOLO. IEEE Access 2023, 11, 102635–102642. [Google Scholar] [CrossRef]
- Koohmishi, M.; Kaewunruen, S.; Chang, L.; Guo, Y. Advancing railway track health monitoring: Integrating GPR, InSAR and machine learning for enhanced asset management. Autom. Constr. 2024, 162, 105378. [Google Scholar] [CrossRef]
- Qiao, X.; Huang, W. A dual frequency transformer network for hyperspectral image classification. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2023, 16, 10344–10358. [Google Scholar] [CrossRef]
- Li, Q.; Ren, S. A visual detection system for rail surface defects. IEEE Trans. Syst. Man Cybern. C 2012, 42, 1531–1542. [Google Scholar] [CrossRef]
- Sresakoolchai, J.; Kaewunruen, S. Railway infrastructure maintenance efficiency improvement using deep reinforcement learning integrated with digital twin based on track geometry and component defects. Sci. Rep. 2023, 13, 2439. [Google Scholar] [CrossRef]
- He, Z.; Wang, Y.; Yin, F.; Liu, J. Surface defect detection for high-speed rails using an inverse PM diffusion model. Sens. Rev. 2016, 36, 86–97. [Google Scholar] [CrossRef]
- Mandriota, C.; Nitti, M.; Ancona, N.; Stella, E.; Distante, A. Filter-based feature selection for rail defect detection. Mach. Vision Appl. 2004, 15, 179–185. [Google Scholar] [CrossRef]
- Augusto Costa, J.; Carmona Cortes, O. A convolutional neural network for detecting faults in power distribution networks along a railway: A case study using YOLO. Appl. Artif. Intell. 2021, 35, 2067–2086. [Google Scholar] [CrossRef]
- Shang, L.; Yang, Q.; Wang, J.; Li, S.; Lei, W. Detection of rail surface defects based on CNN image recognition and classification. In Proceedings of the 2018 20th International Conference on Advanced Communication Technology, Pyeongchang, Republic of Korea, 11–14 February 2018. [Google Scholar] [CrossRef]
- Zhang, H.; Cui, X.; Yin, Y.; Tang, C.; Ding, H.; Zhao, X.; Zhong, J. Comparison and Optimization of Rail Defect Detection Methods Based on Object Detection Model. Tribol. Trans. 2025, 68, 171–179. [Google Scholar] [CrossRef]
- Hamarat, M.; Papaelias, M.; Kaewunruen, S. Fatigue damage assessment of complex railway turnout crossings via Peridynamics-based digital twin. Sci. Rep. 2022, 12, 14377. [Google Scholar] [CrossRef] [PubMed]
- Hussain, M. Yolov1 to v8: Unveiling each variant—A comprehensive review of yolo. IEEE Access 2024, 12, 42816–42833. [Google Scholar] [CrossRef]
- Wang, A.; Chen, H.; Liu, L.; Chen, K.; Lin, Z.; Han, J. Yolov10: Real-time end-to-end object detection. Neural Inf. Process. Syst. 2024, 37, 107984–108011. [Google Scholar] [CrossRef]
- Terven, J.; Córdova-Esparza, D.-M.; Romero-González, J.-A. A comprehensive review of yolo architectures in computer vision: From yolov1 to yolov8 and yolo-nas. Mach. Learn. Knowl. Extr. 2023, 5, 1680–1716. [Google Scholar] [CrossRef]
- Gallagher, J.E.; Oughton, E.J. Surveying you only look once (YOLO) multispectral object detection advancements, applications and challenges. IEEE Access 2025, 13, 7366–7395. [Google Scholar] [CrossRef]
- Upulie, H.; Kuganandamurthy, L. Real-time object detection using YOLO: A review. Sri Lanka Inst. Inf. Technol. Malabe Sri Lanka Tech. Rep. 2021. [Google Scholar] [CrossRef]
- Zhang, Y.; Li, X.; Wang, F.; Wei, B.; Li, L. A comprehensive review of one-stage networks for object detection. In Proceedings of the 2021 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC), Xi’an, China, 17–20 August 2021; pp. 1–6. [Google Scholar] [CrossRef]
- Sehwag, V.; Chiang, M.; Mittal, P. SSD: A Unified Framework for Self-Supervised Outlier Detection. In Proceedings of the International Conference on Learning Representations, Vienna, Austria, 3–7 May 2021. [Google Scholar] [CrossRef]
- Xu, X.; Zhao, M.; Shi, P.; Li, Y. Crack Detection and Comparison Study Based on Faster R-CNN and Mask R-CNN. Sensors 2022, 22, 1215. [Google Scholar] [CrossRef]
- Xu, Y.; Yu, Q.; Wang, Y.; Guo, C.; Feng, S.; Lu, H. Ground target recognition and damage assessment of patrol missiles based on multi-source information fusion. J. Syst. Simul. 2024, 36, 511–521. [Google Scholar] [CrossRef]
- Zhao, Y.; Liu, Z.; Yi, D.; Yu, X.; Sha, X.; Li, L.; Sun, H.; Zhan, Z.; Li, W.J. A review on rail defect detection systems based on wireless sensors. Sensors 2022, 22, 6409. [Google Scholar] [CrossRef]
- Zhang, Z.; Lu, X.; Cao, G.; Yang, Y.; Jiao, L.; Liu, F. ViT-YOLO: Transformer-based YOLO for object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision 2021, Montreal, QC, Canada, 11–17 October 2021. [Google Scholar]
- Li, P.; Zheng, J.; Li, P.; Wang, H. Tomato Maturity Detection and Counting Model Based on MHSA-YOLOv8. Sensors 2023, 23, 6701. [Google Scholar] [CrossRef] [PubMed]
- Xiao, X.; Zhang, D.; Hu, G.; Jiang, Y.; Xia, S. CNN–MHSA: A convolutional neural network and multi-head self-attention combined approach for detecting phishing websites. Neural Netw. 2020, 125, 303–312. [Google Scholar] [CrossRef] [PubMed]
Parameters | Values |
---|---|
Image-size | 640 |
Epochs | 200 |
Batch-size | 16 |
Close-mosaic | 10 |
GPU model | RTX 4090D/24 GB |
Cache | False |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Gao, W.; Geng, R.; Wu, H. Comparative Study on Rail Damage Recognition Methods Based on Machine Vision. Infrastructures 2025, 10, 171. https://doi.org/10.3390/infrastructures10070171
Gao W, Geng R, Wu H. Comparative Study on Rail Damage Recognition Methods Based on Machine Vision. Infrastructures. 2025; 10(7):171. https://doi.org/10.3390/infrastructures10070171
Chicago/Turabian StyleGao, Wanlin, Riqin Geng, and Hao Wu. 2025. "Comparative Study on Rail Damage Recognition Methods Based on Machine Vision" Infrastructures 10, no. 7: 171. https://doi.org/10.3390/infrastructures10070171
APA StyleGao, W., Geng, R., & Wu, H. (2025). Comparative Study on Rail Damage Recognition Methods Based on Machine Vision. Infrastructures, 10(7), 171. https://doi.org/10.3390/infrastructures10070171