Paper contribution

Application of DeepLabV3+ Model in Metro Tunnel Lining Crack Detection

Zhutian Pan; Xuepeng Zhang; Yujing Jiang

Proceedings
EUROCK 2026
Pages
145-150
Submitted
28 January 2026
Accepted
15 June 2026
Published
12 August 2026

Summary

Abstract

Detection of tunnel lining defects is an important part of tunnel maintenance. The manual inspection methods for tunnels are often inefficient and difficult to adapt to a wide range of inspection tasks within the limited maintenance time. To this end, an automatic detection method for metro tunnel lining structures based on deep learning is proposed. The metro tunnel defect detection system (MTDDS) proposed in this paper can capture high-resolution lining surface images with a resolution of 0.2mm/pixel, and the maximum detection speed can reach 20km/h. Furthermore, this paper proposes the EDeepLab model to achieve the detection and segmentation of disease images. Experimental comparisons were conducted on the constructed dataset, and the mean intersection over union (mIoU), class mean pixel accuracy (mPA), and FPS reached 84.77%, 94.96%, and 18.52f/s, respectively. Based on the design method in this paper, the disease detection of the tunnel lining was carried out, and the segmentation result of the algorithm was more stable.

Contributors

Authors

  1. Zhutian Pan Shandong University of Science and Technology, China,
  2. Xuepeng Zhang Shandong University of Science and Technology, China,
  3. Yujing Jiang Shandong University of Science and Technology, China,; Nagasaki Universit, Japan

Index terms

Keywords

Full text

Attachments

Works cited

References

  1. Bang, S. Park, S. Kim, H. & Kim, H. (2019)Encoder–decoder network for pixel‐level road crack detection in black‐box images. Computer‐Aided Civil and Infrastructure Engineering 34(8), 713-727.
  2. Deng, J. Singh, A. Zhou, Y. Lu, Y. & Lee, V. C. S. 2022. Review on computer vision-based crack detection and quantification methodologies for civil structures. Construction and Building Materials 356, 129238.
  3. Gong, Q. Zhu, L. Wang, Y. & Yu, Z. (2021) Automatic subway tunnel crack detection system based on line scan camera. Structural Control and Health Monitoring 28(8), e2776.
  4. Huang, J. Yang, X. Zhou, F. Li, X. Zhou, B. Lu, S. & Slob, E. (2024) A deep learning framework based on improved self‐supervised learning for ground‐penetrating radar tunnel lining inspection. Computer‐Aided Civil and Infrastructure Engineering 39(6), 814-833.
  5. Rafiei, M. H. & Adeli, H. (2016) A novel machine learning model for estimation of sale prices of real estate units. Journal of Construction Engineering and Management 142(2), 04015066.
  6. Rafiei, M. H. & Adeli, H. (2018) Novel machine learning model for construction cost estimation taking into account economic variables and indices. Journal of Construction Engineering and Management 144(12), 04018106.
  7. Ronneberger, O. Fischer, P. & Brox, T. (2015) October. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention pp. 234-241. Cham: Springer international publishing.
  8. Shim, S. (2024) Self‐training approach for crack detection using synthesized crack images based on conditional generative adversarial network. Computer‐Aided Civil and Infrastructure Engineering 39(7), 1019-1041.
  9. Song, G. Huang, Y. Wang, Z. Feng, Q. Jia, F. & Wang, S. (2024) Review of Railway Operation Tunnel Inspection System and Condition Assessment Method. Acta PolytechnicaHungarica 21(1), 259-279.
  10. Wang, Q. Wu, B. Zhu, P. Li, P. Zuo, W. & Hu, Q. (2020) ECA-Net: Efficient channel attention for deep convolutional neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition pp. 11534-11542.
  11. Yu, H. Sun, H. Tao, J. Qin, C. Xiao, D. Jin, Y. & Liu, C. (2023) A multi-stage data augmentation and AD-ResNet-based method for EPB utilization factor prediction. Automation in Construction 147, 104734.
  12. Zhou, Z. Zhang, J. & Gong, C. (2023) Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network. Computer‐Aided Civil and Infrastructure Engineering, 38(17), 2491-2510.