EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Rail surface defect segmentation via strip-aware deformable convolution and spectral deformable transformer

Tong Wang, Qingzhou Mao, Yixuan Shi, Haoxuan Xu, Cuijun Dong, Wei Hu, Zongming Zhang

Transportation Safety and Environment · 10.10.2026 · 8 (3)

DOI: 10.1093/tse/tdag045 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transportation Safety and Environment, online erschienen am 03.08.2026. Verfasst von Tong Wang und 6 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Key Research Program, Natural Science Foundation Project of Hubei Provincial.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate segmentation of rail surface defect regions is critical to the safe operation of railway systems. To achieve precise rail defect segmentation, a segmentation approach that integrates deformable convolution with a spectral deformable transformer (SDT) was proposed. A multi-scale strip-aware orientation-constrained deformable convolution was employed in the upper encoder block, while an SDT was introduced in the deepest encoder block. In the decoder network, a feature-gated fusion module was constructed by combining an Efficient Up-Convolution Block with a Mixed Local Channel Attention module. In the training stage, the pixel-wise classification loss was calculated by a label smoothing weighted cross-entropy function, while the object-level loss was calculated by a weighted generalized Dice function. On this basis, a unified quality score was introduced to enable coordinated optimization of the two independent loss functions. Based on a self-developed railway inspection platform, rail surface defect images were collected from in-service railway lines and a defect segmentation dataset was constructed. Experimental results on the test dataset show that the proposed method achieved 74.89% mIoU and 85.38% F1. These results demonstrate that the proposed framework can accurately and robustly segment slender rail surface defects under complex inspection conditions, showing practical potential for intelligent railway inspection and maintenance.

Förderung

  • National Natural Science Foundation of China
  • National Key Research Program
  • Natural Science Foundation Project of Hubei Provincial
  • Open Research Fund of State Key Laboratory of Rail Transit Vehicle System
  • Southwest Jiaotong University

Zitieren

APA

Wang, T., Mao, Q., Shi, Y., Xu, H., Dong, C., Hu, W., & Zhang, Z. (2026). Rail surface defect segmentation via strip-aware deformable convolution and spectral deformable transformer. Transportation Safety and Environment, 8(3). https://doi.org/10.1093/tse/tdag045

Vancouver

Wang T, Mao Q, Shi Y, Xu H, Dong C, Hu W, et al. Rail surface defect segmentation via strip-aware deformable convolution and spectral deformable transformer. Transportation Safety and Environment. 2026;8(3). doi:10.1093/tse/tdag045

BibTeX
@article{Wang2026,
  author = {Wang, Tong and Mao, Qingzhou and Shi, Yixuan and Xu, Haoxuan and Dong, Cuijun and Hu, Wei and Zhang, Zongming},
  title = {Rail surface defect segmentation via strip-aware deformable convolution and spectral deformable transformer},
  journal = {Transportation Safety and Environment},
  year = {2026},
  volume = {8},
  number = {3},
  doi = {10.1093/tse/tdag045}
}
Kostenlos starten →