Journal Home > Volume 1 , Issue 1

With the rapid development of urban metros, the detection of shield tunnel leakages has become an important research topic. Progressive technological innovations such as deep learning-based methods provide an effective way to detect tunnel leakages accurately and automatically. However, due to the complex shapes and sizes of leakages, it is challenging for existing algorithms to detect such defects. To address these issues, this paper proposes a novel deep learning-based model named segmenting objects by locations network v2 for tunnel leakages (SOLOv2-TL), which is enhanced by ResNeXt-50, deformable convolution, and path augmentation feature pyramid network (PAFPN). In the SOLOv2-TL, ResNeXt-50 coupled with deformable convolution is the backbone for boosting feature extraction ability that would enable the model sensitivity to leakages of different shapes. The PAFPN is introduced as the neck to reduce the loss of leakage information and more accurately assign leakages of different sizes to their corresponding feature levels. The superior performances of ResNeXt-50 with deformable convolution and PAFPN were validated by ablation tests. Moreover, the segmentation results obtained by SOLOv2-TL were compared with those by the mask region-based convolutional neural network (Mask R-CNN), Cascade Mask R-CNN, and SOLO which demonstrated that the mAP, mAP50, and mAP75 of SOLOv2-TL are higher than those of the other methods, where mAP indicates the mean mask average precision (AP) at intersection over union (IoU) = 0.50:0.05:0.95, mAP50 refers to the mean mask AP with an IoU threshold of 0.50, and mAP75 denotes the mean mask AP with an IoU threshold of 0.75. Finally, a leakage area quantification method is presented.


menu
Abstract
Full text
Outline
About this article

Improved SOLOv2 detection method for shield tunnel lining water leakages

Show Author's information Yong Feng1Xiaolei Zhang2Shijin Feng1,2( )Hongxin Chen2Yong Zhao2Yihan Chen1
Urban Mobility Institute, College of Transportation Engineering, Tongji University, Shanghai 201804, China
Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education, Department of Geotechnical Engineering College of Civil Engineering, Tongji University, Shanghai 200092, China

Abstract

With the rapid development of urban metros, the detection of shield tunnel leakages has become an important research topic. Progressive technological innovations such as deep learning-based methods provide an effective way to detect tunnel leakages accurately and automatically. However, due to the complex shapes and sizes of leakages, it is challenging for existing algorithms to detect such defects. To address these issues, this paper proposes a novel deep learning-based model named segmenting objects by locations network v2 for tunnel leakages (SOLOv2-TL), which is enhanced by ResNeXt-50, deformable convolution, and path augmentation feature pyramid network (PAFPN). In the SOLOv2-TL, ResNeXt-50 coupled with deformable convolution is the backbone for boosting feature extraction ability that would enable the model sensitivity to leakages of different shapes. The PAFPN is introduced as the neck to reduce the loss of leakage information and more accurately assign leakages of different sizes to their corresponding feature levels. The superior performances of ResNeXt-50 with deformable convolution and PAFPN were validated by ablation tests. Moreover, the segmentation results obtained by SOLOv2-TL were compared with those by the mask region-based convolutional neural network (Mask R-CNN), Cascade Mask R-CNN, and SOLO which demonstrated that the mAP, mAP50, and mAP75 of SOLOv2-TL are higher than those of the other methods, where mAP indicates the mean mask average precision (AP) at intersection over union (IoU) = 0.50:0.05:0.95, mAP50 refers to the mean mask AP with an IoU threshold of 0.50, and mAP75 denotes the mean mask AP with an IoU threshold of 0.75. Finally, a leakage area quantification method is presented.

Keywords: leakage, metro shield tunnel, instance segmentation, segmenting objects by locations network v2 (SOLOv2)

References(37)

[1]

Y. D. Xue, P. Z. Shi, F. Jia, et al. 3D reconstruction and automatic leakage defect quantification of metro tunnel based on SfM-deep learning method. Undergr Space, 2021, 7: 311–323.

[2]

G. Kuang, B. Li, S. Mo, et al. Review on machine learning-based defect detection of shield tunnel lining. Period Polytech-Civ, 2022, 66: 943–957.

[3]

C. J. Gong, Y. Y. Wang, Y. C. Peng, et al. Three-dimensional coupled hydromechanical analysis of localized joint leakage in segmental tunnel linings. Tunn Undergr SP Tech, 2022, 130: 104726.

[4]

Y. D. Xue, Y. C. Li. A fast detection method via region-based fully convolutional neural networks for shield tunnel lining defects. Comput-Aided Civ Inf, 2018, 33: 638–654.

[5]

Z. Zhou, J. J. Zhang, C. J. Gong. Automatic detection method of tunnel lining multi-defects via an enhanced You Only Look Once network. Comput-Aided Civ Inf, 2022, 37: 762–780.

[6]

Z. Zhou, J. J. Zhang, C. J. Gong, et al. Automatic tunnel lining crack detection via deep learning with generative adversarial network-based data augmentation. Undergr Space, 2023, 9: 140–154.

[7]

S. Zhao, M. Shadabfar, D. M. Zhang, et al. Deep learning-based classification and instance segmentation of leakage-area and scaling images of shield tunnel linings. Struct Control Health Monit, 2021, 28: e2732.

[8]

S. Zhao, D. M. Zhang, Y. D. Xue, et al. A deep learning-based approach for refined crack evaluation from shield tunnel lining images. Automat Constr, 2021, 132: 103934.

[9]
J. Shan, H. D. Cheng, Y. X. Wang. A novel automatic seed point selection algorithm for breast ultrasound images. In: Proceedings of 2008 19th International Conference on Pattern Recognition, Tampa, USA, 2008: pp 1–4.
DOI
[10]

A. Mehnert, P. Jackway. An improved seeded region growing algorithm. Pattern Recogn Lett, 1997, 18: 1065–1071.

[11]

J. P. Fan, D. K. Y. Yau, A. K. Elmagarmid, et al. Automatic image segmentation by integrating color-edge extraction and seeded region growing. IEEE T Image Process, 2001, 10: 1454–1466.

[12]
D. Kumar, A. G. Ramakrishnan. OTCYMIST: Otsu–Canny minimal spanning tree for born-digital images. In: Proceedings of 2012 10th IAPR International Workshop on Document Analysis Systems, Gold Coast, Australia, 2012: pp 389–393.
DOI
[13]

A. S. Ahmed. Comparative study among sobel, Prewitt and Canny edge detection operators used in image processing. J Theor Appl Inf Technol, 2018, 96: 6517–6525.

[14]
J. X. Zhang, W. Chang, L. Wu. Edge detection based on general grey correlation and LoG operator. In: Proceedings of 2010 International Conference on Artificial Intelligence and Computational Intelligence, Sanya, China, 2010: pp 480–483.
DOI
[15]

S. Zhao, D. M. Zhang, H. W. Huang. Deep learning-based image instance segmentation for moisture marks of shield tunnel lining. Tunn Undergr Space Tech, 2020, 95: 103156.

[16]
K. M. He, G. Gkioxari, P. Dollár, et al. Mask R-CNN. In: Proceedings of 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 2017: pp 2980–2988.
[17]

Y. D. Xue, X. Y. Cai, M. Shadabfar, et al. Deep learning-based automatic recognition of water leakage area in shield tunnel lining. Tunn Undergr Space Tech, 2020, 104: 103524.

[18]

Y. Y. Xu, D. W. Li, Q. Xie, et al. Automatic defect detection and segmentation of tunnel surface using modified Mask R-CNN. Measurement, 2021, 178: 109316.

[19]

Y. D. Xue, F. Jia, X. Y. Cai, et al. An optimization strategy to improve the deep learning-based recognition model of leakage in shield tunnels. Comput-Aided Civ Inf, 2022, 37: 386–402.

[20]

S. Liu, H. L. Sun, Z. X. Zhang, et al. A multiscale deep feature for the instance segmentation of water leakages in tunnel using MLS point cloud intensity images. IEEE T Geosci Remote, 2022, 60: 5702716.

[21]
K. M. He, X. Y. Zhang, S. Q. Ren, et al. Deep residual learning for image recognition. In: Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, 2016: pp 770–778.
[22]
T. Y. Lin, P. Dollár, R. Girshick, et al. Feature pyramid networks for object detection. In: Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 2017: pp 936–944.
DOI
[23]

B. C. Russell, A. Torralba, K. P. Murphy, et al. LabelMe: A database and web-based tool for image annotation. Int J Comput Vis, 2008, 77: 157–173.

[24]
X. L. Wang, R. F. Zhang, T. Kong, et al. SOLOv2: Dynamic and fast instance segmentation. In: Proceedings of the 34th International Conference on Neural Information Processing Systems, Vancouver, Canada, 2020: p 1487.
[25]

A. Lara-Doña, S. Torres-Sanchez, B. Priego-Torres, et al. Automated mouse pupil size measurement system to assess locus coeruleus activity with a deep learning-based approach. Sensors, 2021, 21: 7106.

[26]

Y. Zhou, R. G. Cao, P. Li, et al. A damage detection system for inner bore of electromagnetic railgun launcher based on deep learning and computer vision. Expert Syst Appl, 2022, 202: 117351.

[27]
R. Liu, J. Lehman, P. Molino, et al. An intriguing failing of convolutional neural networks and the CoordConv solution. In: Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montréal, Canada, 2018: pp 9628–9639.
[28]
T. Y. Lin, P. Goyal, R. Girshick, et al. Focal loss for dense object detection. In: Proceedings of 2017 IEEE International Conference on Computer Vision, Venice, Italy, 2018: pp 2999–3007.
DOI
[29]
C. H. Sudre, W. Q. Li, T. Vercauteren, et al. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Proceedings of Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Québec City, Canada, 2017: pp 240–248.
DOI
[30]
S. N. Xie, R. Girshick, P. Dollár, et al. Aggregated residual transformations for deep neural networks. In: Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 2017: pp 5987–5995.
DOI
[31]
J. F. Dai, H. Z. Qi, Y. W. Xiong, et al. Deformable convolutional networks. In: Proceedings of 2017 IEEE International Conference on Computer Vision, Venice, Italy, 2017: 764–773.
DOI
[32]
X. Z. Zhu, H. Hu, S. Lin, et al. Deformable ConvNets V2: More deformable, better results. In: Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, USA, 2019: pp 9300–9308.
DOI
[33]
S. Liu, L. Qi, H. F. Qin, et al. Path aggregation network for instance segmentation. In: Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 8759–8768.
DOI
[34]
Open-MMLab/MMDetection: OpenMMLab detection toolbox and benchmark [Online]. https://github.com/open-mmlab/mmdetection (accessed 2022-11-11).
[35]
L. Bottou. Stochastic gradient descent tricks. In: Neural Networks: Tricks of the Trade. G. Montavon, G. B. Orr, K. R. Müller, Eds. Berlin (Germany): Springer, 2012: pp 421–436.
DOI
[36]
Z. W. Cai, N. Vasconcelos. Cascade R-CNN: Delving into high quality object detection. In: Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2017: 6154–6162.
DOI
[37]
X. L. Wang, T. Kong, C. H. Shen, et al. SOLO: Segmenting objects by locations. In: Proceedings of the 16th European Conference on Computer Vision, Springer International Publishing, Glasgow, UK, 2020: pp 649–665.
DOI
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 10 January 2023
Revised: 23 February 2023
Accepted: 24 February 2023
Published: 12 April 2023
Issue date: March 2023

Copyright

© The Author(s) 2023. Published by Tsinghua University Press.

Acknowledgements

Much of the work described in this paper was supported by the National Key R&D Program of China (No. 2020YFC1808105), Science and Technology Commission of Shanghai Municipality (No. 21DZ1204400), and the National Natural Science Foundation of China (No. 42007250). The authors would like to greatly acknowledge all these financial support and express most sincere gratitude.

Rights and permissions

The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Return