All Issue

2023 Vol.39, Issue 7 Preview Page
Detection of Steel Ribs in Tunnel GPR Images Based on YOLO Algorithm

YOLO 알고리즘을 활용한 터널 GPR 이미지 내 강지보재 탐지

Byongkyu Bae1
,
Jaehun Ahn2*
,
Hyunjun Jung3
,
Chang Kyoon Yoo4
배 병규1
,
안 재훈2*
,
정 현준3
,
유 창균4
1Member, Undergraduate Student, Dept. of Civil Engrg., Pusan National Univ.
2Member, Prof., Dept. of Civil and Environmental Engrg., Pusan National Univ.
3Bigdata Team Leader, Dept. of DX, Korea Authority of Land & Infrastructure Safety (KALIS)
4Dept. of Tunnel, Korea Authority of Land & Infrastructure Safety (KALIS)
1정회원, 부산대학교 토목공학과 학사과정
2정회원, 부산대학교 사회환경시스템공학과 정교수
3비회원, 국토안전관리원 디지털기획운영실 빅데이터전략팀 팀장
4비회원, 국토안전관리원 시설안전관리단 터널실
*Corresponding Author
31 July 2023. pp. 31-37
PDF XML Citation
Abstract

Since tunnels are built underground, it is impossible to check visually the location and degree of deterioration of steel ribs. Therefore, in tunnel maintenance, GPR images are generally used to detect steel ribs. While research on GPR image analysis employing artificial neural networks has primarily focused on detecting underground pipes and road damage, there have been limited applications for analyzing tunnel GPR data, specifically for steel rib detection, both internationally and domestically. In this study, a one-step object detection algorithm called YOLO, based on a convolutional neural network, was utilized to automate the localization of steel ribs using GPR data. The performance of the algorithm is then analyzed. Two datasets were employed for the analysis. A dataset comprising 512 original images and another dataset consisting of 2,048 augmented images. The omission rate, which represents the ratio of undetected steel ribs to the total number of steel ribs, was 0.38% for the model using the augmented data, whereas the omission rate for the model using only the original data was 7.18%. Thus, from an automation standpoint, it is more practical to employ an augmented dataset.

Keywords
Convolutional neural network
Data augmentation
Ground penetrating rader
Tunnel maintenance
YOLO

터널은 지중에 건설되는 구조물이므로 육안으로 터널 강지보재의 위치 등의 확인이 불가능하다. 이에, 터널 유지관리시에는, 일반적으로 GPR 이미지를 활용하여 강지보재 탐지를 수행한다. 인공신경망을 통한 GPR 이미지 분석에 대한 연구는, 주로 지하배관, 도로 손상 등의 탐지에 집중되어 있으며, 강지보재 등의 터널 GPR 데이터를 분석한 사례는 해외와 국내 모두 제한적이다. 본 연구에서는, 합성곱 신경망을 기반으로 하는 1단계 객체인식 알고리즘인 YOLO를 활용하여, GPR 데이터를 바탕으로 한 터널 강지보재의 위치 탐지를 자동화하고, 그 성능을 분석한다. 원본 이미지 데이터는 총 512개이며 원본 이미지 데이터로 이루어진 데이터 세트와 원본 이미지 데이터와 증식기법이 적용된 이미지 데이터를 병합한 2,048개의 데이터로 이루어진 데이터 세트를 해석에 활용하였다. 증식한 데이터를 사용한 모델의 강지보재 누락율(전체 강지보재와 탐지하지 못한 지보재 숫자의 비율)은 0.38%, 원본 데이터만을 활용한 모델의 강지보재 누락율은 7.18%로 나타났다. 따라서, 분석 자동화 측면에서는, 증식기법이 적용된 데이터 세트를 활용하는 것이 더 실용적일 것으로 판단된다.

References
1
Girshick, R. (2015), Fast r-cnn. In Proceedings of the IEEE international conference on computer vision (pp.1440-1448). 10.1109/ICCV.2015.169
2
Go, G. H. and Lee, S. J. (2021), A Study on Numerical Analysis for GPR Signal Characterization of Tunnel Lining Cavities, Journal of the Korean Geotechnical Society, Vol.37, No.10, pp.65-76.
3
Go, G. H. and Lee, S. J. (2021), Numerical Modeling for the Identification of Fouling Layer in Track Ballast Ground, Journal of the Korean Geotechnical Society, Vol.37, No.9, pp.13-24.
4
Gu, S., Pednekar, M., and Slater, R. (2019), Improve Image Classification Using Data Augmentation and Neural Networks, SMU Data Science Review, Vol.2, No.2, p.1.
5
He, K., Gkioxari, G., Dollár, P., and Girshick, R. (2017), Mask r-cnn. In Proceedings of the IEEE international conference on computer vision (pp.2961-2969). 10.1109/ICCV.2017.322
6
KALIS (2020), Gyeongbu Hige-Speed Line Cheonan-Asan to Daejeon Unju Tunnel and other facilities Precise Safety Diagnosis Report, Korea.
7
Kim, I.-S., Lee, M. G., and Jeon, Y. (2021), Comparative Analysis of Defect Detection Using YOLO of Deep Learning, Journal of the Korean Society of Manufacturing Technology Engineers, Vol.30, No.6, pp.514-519. https://doi.org/10.7735/ksmte.2021.30.6.51310.7735/ksmte.2021.30.6.513
8
Kim, Kwanghyun, "The collapse of this safety myth… At least 7 killed in tunnel collapse", SBS NEWS, Dec 03, 2012.
9
Ko, H. Y. and Kim, N. G. (2019), Performance Analysis of Detecting Buried Pipelines in GPR Images Using Faster R-CNN, Journal of Convergence for information Technology, Vol.9, No.5, 21-26.
10
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998), Gradient- based Learning Applied to Document Recognition, Proceedings of the IEEE, Vol.86, No.11, pp.2278-2324. 10.1109/5.726791
11
Lee, S. Y., Song, K. I., Kang, K. N., and Ryu, H. H. (2022), Comparison of Performance of Automatic Detection Model of GPR Signal Considering the Heterogeneous Ground, Journal of Korean Tunnelling and Underground Space Association, Vol.24, No.4, pp.341-353.
12
Liang, X., Yu, X., Chen, C., Jin, Y., and Huang, J. (2022), Automatic Classification of Pavement Distress Using 3D Ground- Penetrating Radar and Deep Convolutional Neural Network, IEEE Transactions on Intelligent Transportation Systems, Vol.23, No.11, pp.22269-22277. 10.1109/TITS.2022.3197712
13
Liu, Z., Wu, W., Gu, X., Li, S., Wang, L., and Zhang, T. (2021), Application of Combining YOLO Models and 3D GPR Images in Road Detection and Maintenance, Remote Sensing, Vol.13, No.6, p.1081. 10.3390/rs13061081
14
Ministry of Land, Infrastructure and Transport. (2022), Yearbook of Road Bridge and Tunnel Statistics. (Report No. 11-1613000- 000108-10). Korea Institute of Civil Engineering and Building Technology. https://www.codil.or.kr/viewDtlConRpt.do?pMetaCode= OTKCEC220979&gubun=rpt
15
Özkaya, Umut, et al., "Residual CNN+ Bi-LSTM Model to Analyze GPR B Scan Images", Automation in Construction 123 (2021): 103525. 10.1016/j.autcon.2020.103525
16
Park, C. H., Won, K. S., Byun, J. H., Min, D. H., and Yoon, H. K. (2013), The Investigation of Alluvium by Using Electrical Resistivity, Seismic Survey and GPR, Journal of the Korean Geotechnical Society, Vol.29, No.9, pp.17-29. 10.7843/kgs.2013.29.9.17
17
Park, S., Kim, J., Kim, W., Kim, H., and Park, S. (2019), A Study on the Prediction of Buried Rebar Thickness Using CNN based on GPR Heatmap Image Data, Journal of the Korea institute for structural maintenance and inspection, Vol.23, No.7, pp.66-71.
18
Qin, H., Zhang, D., Tang, Y., and Wang, Y. (2021), Automatic Recognition of Tunnel Lining Elements from GPR Images Using Deep Convolutional Networks with Data Augmentation, Automation in Construction, 130, 103830. 10.1016/j.autcon.2021.103830
19
Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016), You Only Look Once: Unified, Real-time Object Detection, In Proceedings of the IEEE conference on computer vision and pattern recognition (pp.779-788). 10.1109/CVPR.2016.91
20
Seo, K. (2018), Analysis of Evolutionary Optimization Methods for CNN Structures, The Transactions of The Korean Institute of Electrical Engineers, Vol.67, No.6, pp.767-772.
21
Wang, C. Y., Liao, H. Y. M., Wu, Y. H., Chen, P. Y., Hsieh, J. W., and Yeh, I. H. (2020), CSPNet: A New Backbone that Can Enhance Learning Capability of CNN, In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops (pp.390-391). 10.1109/CVPRW50498.2020.00203
Information
  • Publisher :The Korean Geotechnical Society
  • Publisher(Ko) :한국지반공학회
  • Journal Title :Journal of the Korean Geotechnical Society
  • Journal Title(Ko) :한국지반공학회 논문집
  • Volume : 39
  • No :7
  • Pages :31-37
  • Received Date : 2023-05-31
  • Revised Date : 2023-07-10
  • Accepted Date : 2023-07-10
  • DOI :https://doi.org/10.7843/kgs.2023.39.7.31
Journal Informaiton Journal of the Korean Geotechnical Society Journal of the Korean Geotechnical Society
Archives
: Vol.23~28 (2007~2012)
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close