All Issue

2025 Vol.41, Issue 3 Preview Page
A Study on the Application of an AI Model for Joint Detection using NM-GPR

NM-GPR을 활용한 공동 탐지 AI 모델 적용성 연구

Gi-Seok Lee1
,
Jin-Wook Oh2
,
Kang-Ile Lee3*
이 기석1
,
오 진욱2
,
이 강일3*
1Member, Ph.D. Candidate, Dept. of Civil Engineering, Daejin Univ.
2Member, Ph.D. Candidate, Dept. of Civil Engineering, Daejin Univ.
3Member, Prof., Dept. of Civil Engineering, Daejin Univ.
1정회원, 대진대학교 토목공학과 박사과정
2정회원, 대진대학교 토목공학과 박사과정
3정회원, 대진대학교 스마트건설환경공학과 교수
*Corresponding Author
30 June 2025. pp. 15-24
PDF XML Citation
Abstract

This study examines the characteristics of noise-modulated ground penetrating radar (NM-GPR), a device used in GPR-based cavity detection to prevent ground subsidence. It develops an AI-based cavity detection model tailored to its unique features. NM-GPR enhances data visibility through noise modulation, reducing power consumption and enabling rapid and efficient road surveys. Despite its technological advancement, AI-based detection methods for NM-GPR remain underdeveloped. To address this gap, we compare NM-GPR with conventional impulse-based GPR systems and develop an AI model optimized for the distinct characteristics of the NM-GPR. During model development, we applied data preprocessing and augmentation techniques to analyze their impact on detection outcomes and constructed a dedicated training dataset. Using this dataset, we implemented You Only Look Once (YOLO) and faster R-CNN models, evaluating their performance based on mean average precision (mAP), precision, and recall. Finally, we applied the trained models to a test dataset to assess their feasibility for real-world application.

Keywords
AI model
Cavity detection
Faster-RCNN
Noise-modulated ground penetrating radar
Precision
Recall
You Only Look Once

본 연구는 지반 침하 예방을 위해 수행되는 GPR(Ground Penetrating Radar) 공동 탐사에 활용되는 장비 중 NM-GPR(Noise-modulated GPR)의 특성을 분석하고, 이를 기반으로 공동 탐지 모델을 개발하여 적용하는 연구를 수행하였다. NM-GPR은 노이즈를 활용하여 데이터의 시인성을 높이고, 사용 전력을 절감하여 빠른 속도로 도로에서 탐사할 수 있는 장점을 갖고 있다. 그러나 최신 기술임에도 불구하고, 인공지능 기반 탐지 기술의 개발은 미흡한 실정이다. 따라서 본 연구에서는 임펄스 신호를 사용하는 기존 GPR 장비와 NM-GPR을 비교 분석하고, NM-GPR 장비의 특성에 적합한 AI 모델을 적용성을 평가 하였다. AI 모델 적용성 평가를 위해 기존의 데이터 전처리 및 증강 기법을 적용하여 탐지 결과의 변화를 분석하고, 학습 데이터셋을 구축하였다. 구축된 학습 데이터셋을 활용하여 YOLO(You Only Look Once)와 Faster R-CNN 모델을 적용한 후, mAP(mean Average Precision), 정밀도(Precision), 재현율(Recall)을 분석하였다. 또한, 시험 데이터셋을 활용하여 모델의 현장 적용 가능성을 평가하였다.

References
1

Asadi, P., Gindy, M., and Alvarez M. (2019), "A Machine Learning Based Approach for Automatic Rebar Detection and Quantification of Deterioration in Concrete Bridge Deck Ground Penetrating Radar B-scan Images", Journal of Civil Engineering, Vol.23, No.6, pp.2618-2627, https://doi.org/10.1007/s12205-019-2012-z.

10.1007/s12205-019-2012-z
2

Bae, B.K., Ahn, J.H., Jung, H.J., and Yoo, C.K. (2023), "Detection of Steel Ribs in Tunnel GPR Images Based on YOLO Algorithm", Journal of the Korean Geotechnical Society, Vol.39, No.7, pp.31-37, https://doi.org/10.7843/kgs.2023.39.7.31.

3

Bae, Y.S., Kim, K.T., and Lee, S.-Y. (2017), "The Road Subsidence Status and Safety Improvement Plans", Journal of the Korea Academia-Industrial cooperation Society, Vol.18, No.1, pp.545-552, https://doi.org/10.5762/KAIS.2017.18.1.545.

10.5762/KAIS.2017.18.1.545
4

Choi, B.H., Pyun, S.J., Choi, W.C., Choi, C.H., and Yoon, J.S. (2022), "Deep-learning-based GPR Data Interpretation Technique for Detecting Cavities in Urban Roads", Geophysics and Geophysical Explotation, Vol.25, No.4, pp.189-200, https://doi.org/10.7582/GGE.2022.25.4.189.

5

Dinh, K., Gucunski, N., and Duong, Trung H. (2018), "An Algorithm for Automatic Localization and Detection of Rebars from GPR Data of Concrete Bridge Decks", Automation in Construction, Vol.89, pp.292-298, https://doi.org/10.1016/j.autcon.2018.02.017.

10.1016/j.autcon.2018.02.017
6

Girshick, R. (2015), "Fast r-cnn", Proc. IEEE Int. Conf. Comput. vis., Expanded Abstracts, pp.1440-1448, https://doi.org/10.1109/ICCV.2015.169.

10.1109/ICCV.2015.169
7

Han, Y.S. (2018), "Numerical Analysis and Exploring of Ground Condition during Groundwater Drawdown Environment in Open-cut Type Excavation", Journal of the Korean Geothechnical Society, Vol.34, No.11, pp.93-105, https://doi.org/10.7843/kgs.2018.34.11.93.

8

Kim, N.G., Kim, K.D., An, Y.K., Lee, H.J., and Lee, J.J. (2018), "Underground Object Classification Using 3D GPR Data", Korean Society of Road Engineers, p.68, https://db.koreascholar.com/Article/Detail/348241.

9

Reeves, B.A. (2014), "Noise Modulated GPR: Second Generation Technology", 15th International Conference on Ground Penetrating Radar(GPR), pp.708-713, https://doi.org/10.1109/ICGPR.2014.6970519.

10.1109/ICGPR.2014.6970519
10

Reeves, B.A. and Muller, W.B. (2012), "Trafficspeed 3-D Noise Modulated Ground Penetrating Radar(NM-GPR)", 14th International Conference on Ground Penetrating Radar(GPR), pp.165-171, https://doi.org/10.1109/ICGPR.2012.6254854.

10.1109/ICGPR.2012.6254854
11

Ren, S., He, K., Girshick, R., and Sun, J. (2015), "Faster R-CNN: Towards Real-time Object Detection with Region Proposal Networks", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.39, No.6, pp.1137-1149, https://doi.org/10.1109/TPAMI.2016.2577031.

10.1109/TPAMI.2016.257703127295650
12

Thitimakorn, T., Kampananon, N., Jongjaiwanichkit, N., and Kupongsak, S. (2016), "Subsurface Void Detection under the Road Surface Using Ground Penetrating Radar (GPR), a Case Study in the Bangkok Metropolitan Area", Thailand, Int. J. Geotech. Eng., Vol.7, No.1, pp.1-9, https://doi.org/10.1186/s40703-016-0017-8.

10.1186/s40703-016-0017-8
13

Tomita, H., Tada, H., Nanbu, T., Chou, K., Nakamura, T., and McGregor, T., (1995), "Nature and Detection of Void-induced Pavement Failures", Transp. res. Rec., 1505, pp.9-16, https://api.semanticscholar.org/CorpusID:56106050.

14

Tran, V.P. and Lee, H.J. (2020), "Application of Deep Learning Algorithm for Pavement Object Detection Part 1: Asphalt Pavement Crack Detection Part 2: Cavity Detection", Ph. D. Dissertation, The Graduate School of Sejong University, pp.139-147, I804:11042-200000288663.

15

Zou, Z., Shi, Z., Guo, Y., and Ye, J. (2019), "Object Detection in 20 Years : A Survey", arXiv preprint, arXiv:1905.05055, https://doi.org/10.1109/JPROC.2023.3238524.

10.1109/JPROC.2023.3238524
Information
  • Publisher :The Korean Geotechnical Society
  • Publisher(Ko) :한국지반공학회
  • Journal Title :Journal of the Korean Geotechnical Society
  • Journal Title(Ko) :한국지반공학회 논문집
  • Volume : 41
  • No :3
  • Pages :15-24
  • Received Date : 2025-04-16
  • Revised Date : 2025-04-28
  • Accepted Date : 2025-05-08
  • DOI :https://doi.org/10.7843/kgs.2025.41.3.15
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