Hybrid-Supervised-Learning-Based Automatic Image Segmentation for Water Leakage in Subway Tunnels
Source abstract
Abstract
Quickly and accurately identifying water leakage is one of the important components of the health monitoring of subway tunnels. A mobile vision measurement system consisting of several high-resolution, industrial, charge-coupled device (CCD) cameras is placed on trains to implement structural health monitoring in tunnels. Through the image processing technology proposed in this paper, water leakage areas in subway tunnels can be found and repaired in real time. A lightweight automatic segmentation approach to water leakage using hybrid-supervised-deep-learning technology is proposed. This approach consists of the weakly supervised learning Water Leakage-CAM and fully supervised learning WRDeepLabV3+. The Water Leakage-CAM is used for the automatic labeling of data. The WRDeepLabV3+ is used for the accurate identification of water leakage areas in subway tunnels. Compared with other end-to-end semantic segmentation networks, the hybrid-supervised learning approach can more completely segment the water leakage region when dealing with water leakage in complex environments. The hybrid-supervised-deep-learning approach proposed in this paper achieves the highest MIoU of 82.8% on the experimental dataset, which is 6.4% higher than the second. The efficiency is also 25% higher than the second and significantly outperforms other end-to-end deep learning approaches.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Infrastructure Maintenance and Monitoring · Water Systems and Optimization · Concrete Corrosion and Durability
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
Funding
National Natural Science Foundation of China · the postgraduate education and teaching quality improvement project of BUCEA · the BUCEA Post Graduate Innovation Project
References (50)
Show source-provided references
1. Prediction of tunnelling induced ground movement in clay using principle of minimum total potential energy10.1016/j.tust.2022.104854 ↗
2. Experimental investigation of mechanical characteristics for linings of twins tunnels with asymmetric cross-section10.1016/j.tust.2021.104209 ↗
3. Vermeij, D. (2016). Flood Risk Reduction Interventions for the New York City Subway System: A Research on the Impact of Storm Surge and Sea Level Rise on the Safety Against Flooding in Urban Delta’s. [Master’s Thesis, TU Delf].
4. Liu, Y.J. (2019). Research on structural safety and driving dynamic characteristics of Beijing subway shield tunnel under disease. J. Beijing Jiaotong Univ., 1–122.
5. Crack detection and characterization techniques—An overview10.1002/stc.1655 ↗
6. Research progress of machine vision-based disease detecting techniques for the tunnel lining surface
7. A method of disease recognition for shield tunnel lining based on deep learning
8. A high-precision calibration approach for Camera-IMU pose parameters with adaptive constraints of multiple error equations10.1016/j.measurement.2019.107402 ↗
9. A Survey of Indoor Location Technologies, Techniques and Applications in Industry10.1016/j.iot.2022.100608 ↗
10. Wang, X., Wu, Y., Cui, J., Zhu, C.Q., and Wang, X.Z. (2020). Shape characteristics of coral sand from South China Sea. J. Mar. Sci. Eng., 8.10.3390/jmse8100803 ↗
11. Experimental study on mesoscopic shear behavior of calcareous sand material with digital imaging approach10.1155/2020/8881264 ↗
12. Image-based concrete crack detection in tunnels using deep fully convolutional networks10.1016/j.conbuildmat.2019.117367 ↗
13. Real-time tunnel crack analysis system via deep learning10.1109/ACCESS.2019.2916330 ↗
14. A fast detection method via region-based fully convolutional neural networks for shield tunnel lining defects10.1111/mice.12367 ↗
15. Dai, J., Li, Y., He, K., and Sun, J. (2016, January 2–10). R-FCN: Object detection via region-based fully convolutional networks. Proceedings of the 30th International Conference on Neural Information Processing Systems, Red Hook, NY, USA.
16. Deep learning-based image recognition for crack and leakage defects of metro shield tunnel10.1016/j.tust.2018.04.002 ↗
17. Garcia-Garcia, A., Orts-Escolano, S., Oprea, S., Villena-Martinez, V., and Garcia-Rodriguez, J. (2017). A review on deep learning techniques applied to semantic segmentation. arXiv.10.1016/j.asoc.2018.05.018 ↗
18. Multispectral water leakage detection based on a one-stage anchor-free modality fusion network for metro tunnels10.1016/j.autcon.2022.104345 ↗
19. Water leakage image recognition of shield tunnel via learning deep feature representation10.1016/j.jvcir.2019.102708 ↗
20. Patch-based weakly supervised semantic segmentation network for crack detection10.1016/j.conbuildmat.2020.120291 ↗
21. Weakly supervised network based intelligent identification of cracks in asphalt concrete bridge deck10.1016/j.aej.2020.02.027 ↗
22. Chang, Y.T., Wang, Q., Hung, W.C., Piramuthu, R., Tsai, Y.H., and Yang, M.H. (2020, January 13–19). Weakly-supervised semantic segmentation via sub-category exploration. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00901 ↗
23. Deep learning–based image instance segmentation for moisture marks of shield tunnel lining10.1016/j.tust.2019.103156 ↗
24. Pixel-level tunnel crack segmentation using a weakly supervised annotation approach10.1016/j.compind.2021.103545 ↗
25. Automated pixel-level pavement crack detection on 3D asphalt surfaces with a re-current neural network10.1111/mice.12409 ↗
26. Automatic pixel-level multiple damage detection of concrete structure using fully convolutional network10.1111/mice.12433 ↗
27. Autonomous concrete crack detection using deep fully convolutional neural network. Autom10.1016/j.autcon.2018.11.028 ↗
28. Automatic pixel-level crack detection and measurement using fully convolutional network10.1111/mice.12412 ↗
29. Encoder-decoder network for pixel-level road crack detection in black-box images10.1111/mice.12440 ↗
30. Computer vision-based concrete crack detection using U-net fully convolutional networks10.1016/j.autcon.2019.04.005 ↗
31. Mixed supervision for surface- defect detection: From weakly to fully supervised learning10.1016/j.compind.2021.103459 ↗
32. Chen, G.Y., Chen, P.F., Shi, Y.J., Hsieh, C.Y., Liao, B.B., and Zhang, S.Y. (2019). Rethinking the usage of batch normalization and dropout in the training of deep neural networks. arXiv.
33. Deep learning-based sewer defect classification for highly imbalanced dataset10.1016/j.cie.2021.107630 ↗
34. Misra, D. (2019). Mish: A Self Regularized Non-Monotonic Activation Function. arXiv.
35. Zhao, T., and Wu, X.Q. (2019, January 15–20). Pyramid feature attention network for saliency detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.00320 ↗
36. Peng, C., Zhang, X.Y., Yu, G., Luo, G.M., and Sun, J. (2017, January 21–26). Large kernel matters: Improve semantic segmentation by global convolutional network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.189 ↗
37. Focal loss for dense object detection
38. Deep learning-based automatic recognition of water leakage area in shield tunnel lining10.1016/j.tust.2020.103524 ↗
39. Zheng, J.F., Gao, Y.C., Zhang, H., Lei, Y., and Zhang, J. (2022). OTSU Multi-Threshold Image Segmentation Based on Improved Particle Swarm Algorithm. Appl. Sci., 12.10.3390/app122211514 ↗
40. Wu, Y.Y., and Li, Q. (2022). The Algorithm of Watershed Color Image Segmentation Based on Morphological Gradient. Sensors, 22.10.3390/s22218202 ↗
41. Ronneberger, O., Fischer, P., and Brox, T. (2015, January 1–6). U-net: Convolutional net-works for biomedical image segmentation. Proceedings of the International Conference on Medical image computing and computer-assisted intervention, Copenhagen, Denmark.10.1007/978-3-319-24574-4_28 ↗
42. Zhao, H.S., Shi, J.P., Qi, X.J., Wang, X.G., and Jia, J.Y. (2017, January 21–26). Pyramid scene parsing network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.660 ↗
43. Sun, K., Xiao, B., Liu, D., and Wang, J. (2019). Deep high-resolution representation learning for human pose estimation. arXiv.10.1109/CVPR.2019.00584 ↗
44. Chen, L.C., Zhu, Y.K., Papandreou, G., Schroff, F., and Adam, H. (2018, January 1–6). Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European Conference on Computer Vision, Copenhagen, Denmark.10.1007/978-3-030-01234-2_49 ↗
45. Computer vision-based model for moisture marks detection and recognition in subway networks10.1061/(ASCE)CP.1943-5487.0000728 ↗
46. Papandreou, G., Chen, L.C., Murphy, K.P., and Yuille, A.L. (2015, January 7–13). Weakly-and semi-supervised learning of a deep convolutional network for semantic image segmentation. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.10.1109/ICCV.2015.203 ↗
47. Roy, A., and Todorovic, S. (2017, January 21–26). Combining bottom-up, top-down, and smoothness cues for weakly supervised image segmentation. Proceedings of the IEEE Conference on Computer Vision & Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.770 ↗
48. Araslanov, N., and Roth, S. (2020, January 13–19). Single-stage semantic segmentation from image labels. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00431 ↗
49. Zhang, X.R., Peng, Z.L., Zhu, P., Zhang, T.Y., Li, C., Zhou, H.Y., and Jiao, L.C. (2021, January 20–24). Adaptive affinity loss and erroneous pseudo-label refinement for weakly supervised semantic segmentation. Proceedings of the 29th ACM International Conference on Multimedia, Virtual Event.10.1145/3474085.3475675 ↗
50. Screening of Fungi for Potential Application of Self-Healing Concrete10.1038/s41598-019-39156-8 ↗