All Issue

2020 Vol.22, Issue 5 Preview Page

Research Paper

Adversarial learning for underground structure concrete crack detection based on semi-supervised semantic segmentation

지하구조물 콘크리트 균열 탐지를 위한 semi-supervised 의미론적 분할 기반의 적대적 학습 기법 연구

Seungbo Shim1
,
Sang-Il Choi2
,
Suk-Min Kong3
,
Seong-Won Lee4*
심 승보1
,
최 상일2
,
공 석민3
,
이 성원4*
1Research Specialist, Future Infrastructure Research Center, Korea Institute of Civil Engineering and Building Technology
2Senior Researcher, Future Infrastructure Research Center, Korea Institute of Civil Engineering and Building Technology
3Post-Doctoral Researcher, Future Infrastructure Research Center, Korea Institute of Civil Engineering and Building Technology
4Senior Research Fellow, Future Infrastructure Research Center, Korea Institute of Civil Engineering and Building Technology
1정회원, 한국건설기술연구원 차세대 인프라연구센터 전임연구원
2비회원, 한국건설기술연구원 차세대 인프라연구센터 수석연구원
3정회원, 한국건설기술연구원 차세대 인프라연구센터 박사후 연구원
4정회원, 한국건설기술연구원 차세대 인프라연구센터 선임연구위원
* Corresponding Author
30 September 2020. pp. 515-528
PDF XML Citation
Abstract

Underground concrete structures are usually designed to be used for decades, but in recent years, many of them are nearing their original life expectancy. As a result, it is necessary to promptly inspect and repair the structure, since it can cause lost of fundamental functions and bring unexpected problems. Therefore, personnel-based inspections and repairs have been underway for maintenance of underground structures, but nowadays, objective inspection technologies have been actively developed through the fusion of deep learning and image process. In particular, various researches have been conducted on developing a concrete crack detection algorithm based on supervised learning. Most of these studies requires a large amount of image data, especially, label images. In order to secure those images, it takes a lot of time and labor in reality. To resolve this problem, we introduce a method to increase the accuracy of crack area detection, improved by 0.25% on average by applying adversarial learning in this paper. The adversarial learning consists of a segmentation neural network and a discriminator neural network, and it is an algorithm that improves recognition performance by generating a virtual label image in a competitive structure. In this study, an efficient deep neural network learning method was proposed using this method, and it is expected to be used for accurate crack detection in the future.

Keywords
Adversarial learning
Crack detection
Semantic segmentation
Health monitoring
Image processing

통상적으로 콘크리트 지하 구조물은 수십 년 이상 사용할 수 있도록 설계되지만 최근 들어 구조물 중 상당수가 당초의 기대 수명에 근접하고 있는 실정이다. 그 결과 구조물 고유의 기능이 상실되고 다양한 문제가 야기될 수 있어 신속한 점검과 보수가 요구되고 있다. 이를 위해 지금까지는 지하 구조물 유지관리를 위하여 인력 기반의 점검과 보수가 진행되었으나 최근에는 인공지능과 영상 기술의 융합을 통한 객관적인 점검 기술 개발이 활발하게 이루어지고 있다. 특히 딥러닝을 활용한 영상 인식 기술을 적용하여 지도학습 기반의 콘크리트 균열 탐지 알고리즘 개발에 관한 연구가 다양하게 진행되고 있다. 이러한 연구들은 대부분 지도학습 형태 영상 인식 기술로 많은 양의 데이터를 바탕으로 개발이 되는데, 그 중에도 많은 수의 라벨 영상(Label image)이 요구된다. 이를 확보하기 위해서는 현실적으로 많은 시간과 노동력이 필요한 실정이다. 본 논문에서는 이와 같은 문제를 개선하고자 적대적 학습 기법을 적용하여 균열 영역 탐지 정확도를 평균적으로 0.25% 향상시키는 방법을 기술하고자 한다. 이 적대적 학습은 분할(Segmentation) 신경망과 판별자(Discriminator) 신경망으로 구성되어 있고, 가상의 라벨 영상을 경쟁적인 구조로 생성하여 인식 성능을 높이는 알고리즘이다. 본 논문에서는 이 같은 방법을 활용하여 효율적인 심층 신경망 학습 방법을 제시하였고, 향후에 정확한 균열 탐지에 활용될 것으로 기대한다.

키워드
적대적 학습
균열 탐지
의미론적 분할
상태 점검
영상처리
References
1
Arjovsky, M., Chintala, S., Bottou, L. (2017), "Wasserstein generative adversarial networks", Proceedings of the 34th International Conference on Machine Learning, Sydney, pp. 214-223.
2
Cha, Y.J., Choi, W., Büyüköztürk, O. (2017), "Deep learning-based crack damage detection using convolutional neural networks", Computer-Aided Civil and Infrastructure Engineering, Vol. 32, No. 5, pp. 361-378.
10.1111/mice.12263
3
Dawood, T., Zhu, Z., Zayed, T. (2017), "Machine vision-based model for spalling detection and quantification in subway networks", Automation in Construction, Vol. 81, pp. 149-160.
10.1016/j.autcon.2017.06.008
4
Dorafshan, S., Thomas, R.J., Maguire, M. (2018), "SDNET2018: An annotated image dataset for non-contact concrete crack detection using deep convolutional neural networks", Data in Brief, Vol. 21, pp. 1664-1668.
10.1016/j.dib.2018.11.01530505897PMC6247444
5
Feng, C., Zhang, H., Wang, H., Wang, S., Li, Y. (2020), "Automatic pixel-level crack detection on dam surface using deep convolutional network", Sensors, Vol. 20, No. 7, pp. 2069.
10.3390/s2007206932272652PMC7180706
6
FTA (Federal Transit Administration) (1997), "Inspection Policy and Procedures for Rail Transit Tunnels and Underground Structures", TCRP Synthesis 23, Transportation Research Board National Research Council. [Online]. Available:http://onlinepubs.trb.org/onlinepubs/tcrp/tsyn23.pdf
7
He, K., Gkioxari, G., Dollár, P., Girshick, R. (2017), "Mask R-CNN", Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, pp. 2961-2969.
10.1109/ICCV.2017.322
8
Hong, S.H., Kim, J.G., Cho, J.Y., Kim, T.H. (2020), "A study on the necessity of verification and certification system of inspection and diagnostic equipment for infrastructure using advanced technologies", Journal of the Society of Disaster Information, Vol. 16, No. 1, pp. 163-177.
9
Hung, W.C., Tsai, Y.H., Liou, Y.T., Lin, Y.Y., Yang, M.H. (2018), "Adversarial learning for semi-supervised semantic segmentation", arXiv:1802.07934. [Online]. Available: https://arxiv.org/abs/1802.07934
10
Kim, B., Cho, S. (2018), "Automated vision-based detection of cracks on concrete surfaces using a deep learning technique", Sensors, Vol. 18, No. 10, pp. 3452.
10.3390/s1810345230322206PMC6210028
11
Kim, B., Cho, S. (2019), "Image-based concrete crack assessment using mask and region-based convolutional neural network", Structural Control and Health Monitoring, Vol. 26, No. 8, pp. e2381.
10.1002/stc.2381
12
Kim, M.K., Sohn, H., Chang, C.C. (2015), "Localization and quantification of concrete spalling defects using terrestrial laser scanning", Journal of Computing in Civil Engineering, Vol. 29, No. 6, pp. 04014086.
10.1061/(ASCE)CP.1943-5487.0000415
13
Kingma, D.P., Ba, J. (2015), "ADAM: a method for stochastic optimization", Proceedings of the 3rd International Conference on Learning Representations (ICLR), San Diego, pp. 1-15.
14
Kohavi, R. (1995), "A study of cross-validation and bootstrap for accuracy estimation and model selection", Proceedings of the 14th International Joint Conference on Artificial Intelligence, Montreal, Vol. 14, No. 2, pp. 1137-1145.
15
Lee, Y.I., Kim, B., Cho, S. (2018), "Image-based spalling detection of concrete structure using deep learning", Journal of the Korea Concrete Institute, Vol. 30, No. 1, pp. 91-99.
10.4334/JKCI.2018.30.1.091
16
Long, J., Shelhamer, E., Darrell, T. (2015), "Fully convolutional networks for semantic segmentation", Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, pp. 3431-3440.
10.1109/CVPR.2015.7298965
17
Marszalek, M., Schmid, C. (2007), "Accurate object localization with shape masks", Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Minneapolis, pp. 1-8.
10.1109/CVPR.2007.383085
18
O'Byrne, M., Ghosh, B., Schoefs, F., Pakrashi, V. (2014), "Regionally enhanced multiphase segmentation technique for damaged surfaces", Computer-Aided Civil and Infrastructure Engineering, Vol. 29, No. 9, pp. 644-658.
10.1111/mice.12098
19
Park, C.H., Lee, H.I. (2016), Future trend of capital investment for Korean transportation infrastructure, Construction and Economy Research Institute of Korea, pp. 6-8.
20
Pohlen, T., Hermans, A., Mathias, M., Leibe, B. (2017), "Full-resolution residual networks for semantic segmentation in street scenes", Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, pp. 4151-4160.
10.1109/CVPR.2017.353
21
Radford, A., Metz, L., Chintala, S. (2015), "Unsupervised representation learning with deep convolutional generative adversarial networks", arXiv:1511.06434, [Online], Available: https://arxiv.org/abs/1511.06434
22
Ruder, S. (2016), "An overview of gradient descent optimization algorithms", arXiv:1609.04747. [Online]. Available: https://arxiv.org/abs/1609.04747
23
ScrapeBox-The Swiss Army Knife of SEO! Available online: http://www.scrapebox.com/ (accessed on 29 June 2020).
24
Xu, B., Wang, N., Chen, T., Li, M. (2015), "Empirical evaluation of rectified activations in convolutional network", arXiv:1505.00853, [Online], Available: https://arxiv.org/abs/1505.00853
25
Yu, S.N., Jang, J.H., Han, C.S. (2007), "Auto inspection system using a mobile robot for detecting concrete cracks in a tunnel", Automation in Construction, Vol. 16, No. 3, pp. 255-261.
10.1016/j.autcon.2006.05.003
26
Zhang, W., Zhang, Z., Qi, D., Liu, Y. (2014), "Automatic crack detection and classification method for subway tunnel safety monitoring", Sensors, Vol. 14, No. 10, pp. 19307-19328.
10.3390/s14101930725325337PMC4239952
Information
  • Publisher :Korean Tunneling and Underground Space Association
  • Publisher(Ko) :한국터널지하공간학회
  • Journal Title :Journal of Korean Tunnelling and Underground Space Association
  • Journal Title(Ko) :한국터널지하공간학회 논문집
  • Volume : 22
  • No :5
  • Pages :515-528
  • Received Date : 2020-07-21
  • Revised Date : 2020-08-04
  • Accepted Date : 2020-08-04
  • DOI :https://doi.org/10.9711/KTAJ.2020.22.5.515
Journal Informaiton Journal of Korean Tunnelling and Underground Space Association Journal of Korean Tunnelling and Underground Space Association
Online Submission & Review System
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close