Journal of Korean Tunnelling and Underground Space Association (한국터널지하공간학회 논문집)

Korean Tunnelling and Underground Space Association (한국터널지하공간학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of performance of automatic detection model of GPR signal considering the heterogeneous ground

지반의 불균질성을 고려한 GPR 신호의 자동탐지모델 성능 비교

  • Lee, Sang Yun (Dept. of Civil Engineering, Inha University) ;
  • Song, Ki-Il (Dept. of Civil Engineering, Inha University) ;
  • Kang, Kyung Nam (Research Institute of Construction & Environmental System, Inha University) ;
  • Ryu, Hee Hwan (Structural & Seismic Technology Group, Korea Electric Power Corporation Research Institute)
  • 이상연 (인하대학교 토목공학과) ;
  • 송기일 (인하대학교 토목공학과) ;
  • 강경남 (인하대학교 건설환경시스템연구소) ;
  • 류희환 (한국전력공사 전력연구원 구조내진연구실)
  • Received : 2022.06.28
  • Accepted : 2022.07.18
  • Published : 2022.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

Pipelines are buried in urban area, and the position (depth and orientation) of buried pipeline should be clearly identified before ground excavation. Although various geophysical methods can be used to detect the buried pipeline, it is not easy to identify the exact information of pipeline due to heterogeneous ground condition. Among various non-destructive geo-exploration methods, ground penetration radar (GPR) can explore the ground subsurface rapidly with relatively low cost compared to other exploration methods. However, the exploration data obtained from GPR requires considerable experiences because interpretation is not intuitive. Recently, researches on automated detection technology for GPR data using deep learning have been conducted. However, the lack of GPR data which is essential for training makes it difficult to build up the reliable detection model. To overcome this problem, we conducted a preliminary study to improve the performance of the detection model using finite difference time domain (FDTD)-based numerical analysis. Firstly, numerical analysis was performed with homogeneous soil media having single permittivity. In case of heterogeneous ground, numerical analysis was performed considering the ground heterogeneity using fractal technique. Secondly, deep learning was carried out using convolutional neural network. Detection Model-A is trained with data set obtained from homogeneous ground. And, detection Model-B is trained with data set obtained from homogeneous ground and heterogeneous ground. As a result, it is found that the detection Model-B which is trained including heterogeneous ground shows better performance than detection Model-A. It indicates the ground heterogeneity should be considered to increase the performance of automated detection model for GPR exploration.

도심지에는 많은 지중 매설관이 설치되어 있으며, 이러한 지중 관로의 위치(깊이, 방향 등)은 굴착을 수행하기 전에 특정되어야 한다. 지중 매설관을 탐지하기 위해 다양한 지구물리학적인 방법을 사용할 수 있으나, 지반의 불균질성으로 인해 정확한 위치정보를 파악하는 것은 어렵다. 다양한 비파괴 탐사 방법 중 GPR (ground penetrating radar)는 고속으로 실험이 가능하며, 다른 탐사 방법에 비해 상대적으로 저렴한 탐사비용 등의 장점을 갖는다. 그러나 GPR의 탐사 데이터는 해석이 직관적이지 않아 상당한 전문적 지식이 요구된다. 최근 딥러닝을 이용한 탐사 데이터의 자동판독 기술에 대한 연구가 증가하고 있으나, 매설물의 위치를 정확히 알고 있는 탐사 데이터가 부족하여 학습모델 구축에 어려움이 있다. 이를 해결하기 위해 본 연구에서는 이러한 문제를 FDTD (finite difference time domain)수치해석을 통해 해결하고 자동탐지 학습 모델의 성능을 향상시키기 위한 기초연구를 수행하였다. 첫째, 단일유전율로 구성된 균질지반을 구성하고 해석을 수행하였다. 불균질 지반의 경우 프랙탈 기법을 이용하여 모델을 구성하고 해석을 수행하였다. 둘째, 합성곱 신경망을 이용하여 딥러닝 학습을 수행하였다. Model-A는 균질 지반 해석 데이터만 이용하여 학습을 수행하였으며, Model-B는 균질 및 불균질 지반 해석 데이터를 이용하여 학습을 수행하였다. 그 결과 Model-B가 Model-A보다 탐지성능이 우수한 것을 확인하였다. 이는 자동탐지 모델의 학습 시, 지반의 불균질성을 포함하여 학습을 수행하면 탐지 모델의 성능이 개선됨을 의미한다.

Keywords

Acknowledgement

본 연구는 한국전력공사 자체연구개발과제(R21SA02)와 기초연구과제(R21XO01-47)의 지원을 받았습니다. 이에 감사드립니다.

References

  1. Al-Nuaimy, W., Huang, Y., Nakhkash, M., Fang, M.T.C., Nguyen, V.T., Eriksen, A. (2000), "Automatic detection of buried utilities and solid objects with GPR using neural networks and pattern recognition", Journal of applied Geophysics, Vol. 43, No. 2-4, pp. 157-165. https://doi.org/10.1016/S0926-9851(99)00055-5
  2. Benedetto, A., Pajewski, L. (2015), Civil Engineering Applications of Ground Penetrating Radar, Springer, London, pp. i-xi.
  3. Bochkovskiy, A., Wang, C.Y., Liao, H.Y.M. (2020), "Yolov4: Optimal speed and accuracy of object detection", arXiv preprint arXiv:2004.10934, pp. 1-17.
  4. Chae, J.H., Ko, H.Y., Lee, B.G., Kim, N.G. (2019), "A study on the pipe position estimation in GPR images using deep learning based convolutional neural network", Journal of Internet Computing and Services, Vol. 20, No. 4, pp. 39-46.
  5. Fang, Y., Guo, X., Chen, K., Zhou, Z., Ye, Q. (2021), "Accurate and automated detection of surface knots on sawn timbers using YOLO-V5 model", BioResources, Vol. 16, No. 3, pp. 5390-5406. https://doi.org/10.15376/biores.16.3.5390-5406
  6. Giannakis, I. (2016), Realistic numerical modelling of ground penetrating radar for landmine detection, Ph.D. Thesis, University of Edinburgh, pp. 1-268.
  7. Giannopoulos, A. (1998), The investigation of transmission-line matrix and finite-difference time-domain methods for the forward problem of ground probing radar, Ph.D. Thesis, University of York, pp. 1-258.
  8. Giannopoulos, A. (2005), "Modelling ground penetrating radar by GprMax", Construction and Building Materials, Vol. 19, No. 10, pp. 755-762. https://doi.org/10.1016/j.conbuildmat.2005.06.007
  9. Kim, H.M., Bae, H.R. (2021), "A study on GPR image classification by semi-supervised learning with CNN", The Journal of Bigdata, Vol. 6, No. 1, pp. 197-206.
  10. Lee, D.Y. (2018), "Analysis of sewer pipe defect and ground subsidence risk by using CCTV and GPR monitering results", Journal of the Korean Geosynthetics Society, Vol. 17, No. 3, pp. 47-55. https://doi.org/10.12814/JKGSS.2018.17.3.047
  11. Peplinski, N.R., Ulaby, F.T., Dobson, M.C. (1995), "Dielectric properties of soils in the 0.3-1.3-GHz range", IEEE Transactions on Geoscience and Remote Sensing, Vol. 33, No. 3, pp. 803-807. https://doi.org/10.1109/36.387598
  12. Pham, M.T., Lefevre, S. (2018), "Buried object detection from B-scan ground penetrating radar data using Faster-RCNN", Proceedings of the IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, pp. 6804-6807.
  13. Redmon, J., Divvala, S., Girshick, R., Farhadi, A. (2016), "You only look once: Unified, real-time object detection", Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, pp. 779-788.
  14. Taflove, A., Hagness, S.C., Piket-May, M. (2005), Computational Electrodynamics: The Finite-Difference Time-Domain Method, The Electrical Engineering Handbook, 3, Elsevier, Burlington, pp. 629-670.
  15. Tian, Y., Yang, G., Wang, Z., Wang, H., Li, E., Liang, Z. (2019), "Apple detection during different growth stages in orchards using the improved YOLO-V3 model", Computers and Electronics in Agriculture, Vol. 157, pp. 417-426. https://doi.org/10.1016/j.compag.2019.01.012
  16. Warren, C., Giannopoulos, A. gprMax user guide, https://docs.gprmax.com/en/latest/ (Apr 14, 2022)
  17. Warren, C., Giannopoulos, A., Giannakis, I. (2016), "gprMax: Open source software to simulate electromagnetic wave propagation for Ground Penetrating Radar", Computer Physics Communications, Vol. 209, pp. 163-170. https://doi.org/10.1016/j.cpc.2016.08.020
  18. Yuan, C., Li, S., Cai, H., Kamat, V.R. (2018), "GPR signature detection and decomposition for mapping buried utilities with complex spatial configuration", Journal of Computing in Civil Engineering, Vol. 32, No. 4, pp. 1-15.