To address the issues of low efficiency, high missed detection rates, and significant interference from environmental lighting in manual inspection of high-speed railway rail surface damage, this paper designs a rail damage detection system based on artificial intelligence (AI) vision technology. In terms of hardware system construction, the system selects a Huarui Technology (iRAYPLE) 6-megapixel high-resolution industrial camera paired with a Microvision low-distortion high-definition industrial fixed-focus lens to ensure clear imaging of minute defect features. In response to the reflective characteristics of the rail surface and complex on-site lighting conditions, a high-density LED ring light source system is specially designed, which employs uniform cold white light irradiation to effectively suppress shadows and highlight defect edges. Building upon this, advanced deep learning algorithms are integrated to process the captured images, enabling the automated identification of rail surface damages such as cracks, spalling, and abrasions. The research demonstrates that through the synergistic optimization of software and hardware, this system effectively overcomes the limitations of traditional detection methods, significantly improving the accuracy and efficiency of rail damage detection, thereby providing reliable intelligent technical support for railway maintenance.
References
[1] Wang, Y., Miao, B., Zhang, Y., Huang, Z., Xu, S. (2025) Review on rail damage detection technologies for high-speed trains. Applied Sciences, 15(14), 7725.
[2] Luo, H., Cai, L., Li, C. (2023) Rail surface defect detection based on an improved YOLOv5s. Applied Sciences, 13(12), 7330.
[3] Aguirre, E., García-Silvente, M. (2023) Detecting and tracking using 2D laser range finders and deep learning. Neural Computing and Applications, 35(1), 415-428.
[4] Huang, Z., Fu, H. L., Fan, X. D., Meng, J. H., Chen, W., Zheng, X. J., Zhang, J. B. (2021) Rapid surface damage detection equipment for subway tunnels based on machine vision system. Journal of Infrastructure Systems, 27(1), 04020047.
[5] Choi, J. Y., Han, J. M. (2024) Deep learning (Fast R-CNN)-based evaluation of rail surface defects. Applied Sciences, 14(5), 1874.
[6] Saiz, F. A., Barandiaran, I., Arbelaiz, A., Graña, M. (2022) Photometric stereo-based defect detection system for steel components manufacturing using a deep segmentation network. Sensors, 22(3), 882.
[7] Kumar, A., Harsha, S. P. (2025) A systematic literature review of defect detection in railways using machine vision-based inspection methods. International Journal of Transportation Science and Technology, 18, 207-226.
[8] Wang, H., Li, M., Wan, Z. (2022) Rail surface defect detection based on improved Mask R-CNN. Computers and Electrical Engineering, 102, 108269.
[9] Bai, T., Fu, H., Xu, G., Xin, X., Duan, J., Zong, H. (2025) Crack-YOLOv8: multi-class crack detection method for track slabs based on improved YOLOv8. Measurement Science and Technology, 36(5), 055012.
[10] Mi, Z., Chen, R., Zhao, S. (2023) Research on steel rail surface defects detection based on improved YOLOv4 network. Frontiers in Neurorobotics, 17, 1119896.
Share and Cite
Sun G., Sun Q., Chen X., Ling H., Wang Y. (2025) Design of a Rail Damage Detection System Based on AI Vision Technology. Scientific Research Bulletin, 2(6), 120-125.