Automatic detection of tunnel lining cracks has been widely studied for infrastructure safety, but reliable recognition in real operating tunnels remains challenging. Visual interference from glare, shadows, seepage traces, segment joints, and textured backgrounds can obscure fine or discontinuous cracks, making missed detection a major practical risk.
In a study published in the Journal of Railway Science and Technology , a team of researchers proposed an enhanced U-Net architecture for tunnel lining crack segmentation known as CSwinUKAN.
“The model combines a pretrained Swin Transformer to capture long-range crack context, a Tokenized Kolmogorov-Arnold Network module to strengthen nonlinear feature representation, and Swin-based decoding to improve crack continuity recovery,” explains co-corresponding author
Da Shi. “Focal Loss is further adopted to address the imbalance between crack and background pixels and reduce false negatives.”
Validated on an integrated dataset of 10,422 images from public crack datasets and metro shield-tunnel images from Ningbo and Shanghai, CSwinUKAN outperformed U-Net, Swin-U-Net, U-KAN, and YOLO-based baselines, achieving stronger segmentation performance and recall.
“This provides a promising approach for high-accuracy tunnel crack inspection and lays a foundation for future lightweight and real-time deployment,” says co-correspond author Xuguang Wang.
The researchers encouraged subsequent research to focus on optimizing computational efficiency. While the computational demand of the current model is feasible for static image segmentation utilizing contemporary industrial-grade GPUs, it spells a functional limitation for seamless integration into fully automated, real-time diagnostic systems.
###
Contact the author: Xuguang Wang, University of Hong Kong, xuguangw@hku.hk .
The publisher KeAi was established by Elsevier and China Science Publishing & Media Ltd to unfold quality research globally. In 2013, our focus shifted to open access publishing. We now proudly publish more than 200 world-class, open access, English language journals, spanning all scientific disciplines. Many of these are titles we publish in partnership with prestigious societies and academic institutions, such as the National Natural Science Foundation of China (NSFC).
Journal of Railway Science and Technology
CSwinUKAN: An enhanced U-Net architecture with swin-transformer and Kolmogorov-Arnold network modification for tunnel lining cracks segmentation.
All authors declare that there are no competing interests.