Journal of the Korea Society of Computer and Information (한국컴퓨터정보학회논문지)

Korean Society of Computer Information (한국컴퓨터정보학회)
권호 표지
DOI QR코드

DOI QR Code

Automatic Detection System of Underground Pipe Using 3D GPR Exploration Data and Deep Convolutional Neural Networks

  • Son, Jeong-Woo (Dept. of Computer Science and Engineering, Kangwon National University) ;
  • Moon, Gwi-Seong (Dept. of Computer Science and Engineering, Kangwon National University) ;
  • Kim, Yoon (Dept. of Computer Science and Engineering, Kangwon National University)
  • Received : 2020.11.20
  • Accepted : 2021.01.18
  • Published : 2021.02.26
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose Automatic detection system of underground pipe which automatically detects underground pipe to help experts. Actual location of underground pipe does not match with blueprint due to various factors such as ground changes over time, construction discrepancies, etc. So, various accidents occur during excavation or just by ageing. Locating underground utilities is done through GPR exploration to prevent these accidents but there are shortage of experts, because GPR data is enormous and takes long time to analyze. In this paper, To analyze 3D GPR data automatically, we use 3D image segmentation, one of deep learning technique, and propose proper data generation algorithm. We also propose data augmentation technique and pre-processing module that are adequate to GPR data. In experiment results, we found the possibility for pipe analysis using image segmentation through our system recorded the performance of F1 score 40.4%.

본 논문에서는 관로를 자동으로 검출하는 지하 관로 자동 탐색 시스템을 제안한다. 시간에 따른 지반변화, 관로 시공 불일치 등 여러 가지 요인으로 실제 관로의 위치가 지하 관로 도면과 일치하지 않는다. 이로 인하여 굴착공사나 관로 노후화에 의한 여러 사고가 발생한다. 사고를 방지하기 위해 GPR(지표 투과 레이더, Ground Penetrating Radar) 탐사를 통해 지하시설물을 찾아내는 작업이 이루어지고 있지만, 분석을 담당할 수 있는 전문가의 수가 부족하다. GPR 데이터는 매우 방대하며 분석과정에도 오랜 시간이 걸리기 때문이다. 이에 본 논문에서는 3D GPR 데이터를 자동으로 분석하기 위해 딥 러닝 기술인 3D 이미지 분할을 사용하고, 이에 적합한 데이터 생성 알고리즘을 제안한다. 또한 GPR 데이터 특성에 맞는 데이터 증강 기법, 데이터 전처리 모듈을 제안한다. 실험 결과를 통해 제안한 시스템은 F1 Score 40.4%의 성능을 보였으며 이를 통해 이미지 분할을 이용한 관로 분석의 가능성을 확인하였다.

Keywords

References

  1. D. Y. Lee, "Analysis of Sewer Pipe Defect and Ground Subsidence Risk By Using CCTV and GPR Monitering Results", J. Korean Geosynthetics Society, Vol. 17, No. 3, pp. 47-55, Sep. 2018. DOI: http://dx.doi.org/10.12814/jkgss.2018.17.3.047
  2. Daniels, David J. "Ground penetrating radar.", John Wiley & Sons, Inc, 2005.
  3. Lee, JaeWon, DoHyeon Kim, and Yoon Kim. "Online Video Synopsis via Multiple Object Detection.", Journal of the Korea Society of Computer and Information, Vol. 24, No. 8, pp. 19-28, Aug. 2019. DOI: https://doi.org/10.9708/jksci.2019.24.08.019
  4. Lee, Dong-Ryeol, and Yoon Kim. "Multi-scale face detector using anchor free method." Journal of the Korea Society of Computer and Information, Vol. 25, No. 7, pp. 47-55, Jul. 2020. DOI: https://doi.org/10.9708/jksci.2020.25.07.047
  5. J. Chae, H. Ko, B. Lee, and N. Kim, "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, Aug. 2019. DOI: https://doi.org/10.7472/jksii.2019.20.4.39
  6. H.-Y. Ko and N. Kim, "Performance Analysis of Detecting buried pipelines in GPR images using Faster R-CNN", Journal of Convergence for Information Technology, Vol. 9, No. 5, pp. 21-26, May 2019. DOI: https://doi.org/10.22156/CS4SMB.2019.9.5.021
  7. Kim, Y. T., Kim, B., Kim, J. W. Park, H. M. , and Yoon, J. S. "Determining the Optimal Frequency of Ground Penetrating Radar for Detecting Voids in Pavements", Int. J. Highw. Eng, Vol. 18, No. 2, pp. 37-42, Mar. 2016. DOI: https://doi.org/10.7855/IJHE.2016.18.2.037
  8. Bianchini Ciampoli, Luca, et al. "Signal processing of GPR data for road surveys." Geosciences, Vol. 9, No. 2, pp. 96, Feb. 2019. DOI: https://doi.org/10.3390/geosciences9020096
  9. Park, Byeongjin, et al. "Underground object classification for urban roads using instantaneous phase analysis of Ground-Penetrating Radar (GPR) Data." Remote Sensing, Vol. 10, No. 9, pp. 1417, Sep. 2018. DOI: https://doi.org/10.3390/rs10091417
  10. Kim, Namgyu, et al. "Deep learning-based underground object detection for urban road pavement." International Journal of Pavement Engineering, Vol. 21, No. 13, pp. 1638-1650, Dec. 2020. DOI: https://doi.org/10.1080/10298436.2018.1559317
  11. Dinh, Kien, Nenad Gucunski, and Trung H. Duong. "An algorithm for automatic localization and detection of rebars from GPR data of concrete bridge decks." Automation in Construction, Vol. 89, pp. 292-298, May. 2018. DOI: https://doi.org/10.1016/j.autcon.2018.02.017
  12. Long, Jonathan, Evan Shelhamer, and Trevor Darrell. "Fully convolutional networks for semantic segmentation." In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 3431-3440, Mar. 2015.
  13. Cicek, Ozgun, et al. "3D U-Net: learning dense volumetric segmentation from sparse annotation." International conference on medical image computing and computer-assisted intervention, pp. 424-432, Jun. 2016. DOI: https://doi.org/10.1007/978-3-319-46723-8_49
  14. Huang, Yi-Jie, et al. "3-D RoI-Aware U-Net for Accurate and Efficient Colorectal Tumor Segmentation." IEEE Transactions on Cybernetics, Feb. 2020. DOI: https://doi.org/10.1109/tcyb.2020.2980145
  15. Shorten, Connor, and Taghi M. Khoshgoftaar. "A survey on image data augmentation for deep learning." Journal of Big Data, Vol. 6, No.1, pp. 60, Jul. 2019. DOI: https://doi.org/10.1186/s40537-019-0197-0
  16. Hyun, Seung-Yeup. "Characteristic Changes in Ground-Penetra ting Radar Responses from Dielectric-Filled Nonmetallic Pipes Buried in Inhomogeneous Ground.", The Journal of Korean Institute of Electromagnetic Engineering and Science, Vol. 30, No. 5, pp. 399-406, May 2019. DOI: https://doi.org/10.5515/KJKIEES.2019.30.5.399
  17. Dice, Lee R. "Measures of the amount of ecologic association between species.", Ecology, Vol. 26, No. 3, pp. 297-302, Jul. 1945. DOI: https://doi.org/10.2307/1932409

Cited by

  1. SKU-Net: Improved U-Net using Selective Kernel Convolution for Retinal Vessel Segmentation vol.26, pp.4, 2021, https://doi.org/10.9708/jksci.2021.26.04.029