한국국방기술학회 논문지 (Journal of The Korean Institute of Defense Technology)

한국국방기술학회 (The Korean Institute of Defense Technology)
권호 표지

시뮬레이션 데이터 기반으로 학습된 딥러닝 모델을 활용한 지뢰식별연구

Deep-Learning-Based Mine Detection Using Simulated Data

  • 전부환 (국방대학교 국방과학 무기체계학과) ;
  • 이춘주 (국방대학교 국방과학)
  • 투고 : 2023.11.24
  • 심사 : 2023.12.29
  • 발행 : 2023.12.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

세계적으로 지뢰의 수는 감소하는 추세이지만, 과거에 묻힌 지뢰로 인한 피해는 계속되고 있다. 이에따라 본 연구는 지뢰탐지 장비의 개선과 미래 군인 수의 감소 등으로 인해 발생할 수 있는 문제점, 제한사항에 대한 해결방안을 생각하였다. 현재 지뢰탐지기들에는 데이터 저장 기능이 탑재되어 있지 않아 연구 등을 위한 데이터 구축에 제한사항이 있다. 그리고 실제 환경에서 데이터 구축은 많은 시간과 인력이 들어가게된다. 그래서 본 연구에서는 gprMax 시뮬레이션을 활용하여 데이터를 생성하고, CNN 기반의 경량 모델인 MobileNet을 학습시켰고, 실제 데이터로 검증한 결과 97.35%의 높은 식별율을 볼 수 있었다. 그러므로 딥러닝, 시뮬레이션 등의 기술이 지리탐지 장비 등에 접목되는 가능성을 보고, 미래 발생할 수 있을 문제점을 어느정도 해소하고 우리군이 미래 과학기술군이 되기위한 무기체계 발전의 발판이 되길 기대한다.

Although the global number of landmines is on a declining trend, the damages caused by previously buried landmines persist. In light of this, the present study contemplates solutions to issues and constraints that may arise due to the improvement of mine detection equipment and the reduction in the number of future soldiers. Current mine detectors lack data storage capabilities, posing limitations on data collection for research purposes. Additionally, practical data collection in real-world environments demands substantial time and manpower. Therefore, in this study, gprMax simulation was utilized to generate data. The lightweight CNN-based model, MobileNet, was trained and validated with real data, achieving a high identification rate of 97.35%. Consequently, the potential integration of technologies such as deep learning and simulation into geographical detection equipment is highlighted, offering a pathway to address potential future challenges. The study aims to somewhat alleviate these issues and anticipates contributing to the development of our military capabilities in becoming a future scientific and technological force.

키워드

참고문헌

  1. C Warren, A Giannopoulos, I Giannakis " gprMax : Open source software to simulate electromagnetic wave propagation for Ground Penetrating Radar", 「Computer Physics」, 2016
  2. Silvia Lameri, Federico Lombardi "Landmine Detection from GPR Data Using Convolutional Neural Networks", 「IEEE Xplore」, 2076-1465, 2017
  3. Venceslav Kafedziski, Senior Member, "Detection and Classification of Land Mines from Ground Penetrating Radar Data Using Faster R-CNN", 「IEEE Xplore」, 2018
  4. Xianyang Gao, Frank JW Podd, Wouter van Verre, David J Daniels, Yee M Tan; "Simulation of Ground Penetrating Radar for Anti-personnel Landmine Detection", 「IEEE Xplore」, 2474-3844, 2018
  5. (LeCun et al, "Deep learning", 「nature」, 521, 7553(2015): 436-444. https://doi.org/10.1038/nature14539
  6. Ham, J., Kim, M., Heo, E., Kim, S., & Choi, S. (2017). "Portable Mine Detection System Based on Ground Penetrating Radar with Frequency Shift Filtering." Journal of the Institute of Electronics Engineers of Korea, Vol. 54, No. 5.
  7. Chae, C., Ko, H., Lee, B., & Kim, N. (2019). "A Study on Pipeline Location Estimation using Deep Learning-based CNN in GPR Images." Journal of Internet Computing and Services.
  8. Yoo, Y., Kim, D., Lee, M., & Lee, J. (2021). "Deep Neural Network Learning Based on Simulation-generated Data for Underground Bunker Detection." Journal of the Korean Institute of Communication Sciences, Vol. 46, No. 05, pp. 21-05.
  9. Kim, D., & Choi, S. (2021). "Signal Processing Research for Array-type Broadband Ground Penetrating Radar-based Mine Detection." Journal of the Institute of Electronics Engineers of Korea, Vol. 58, No. 9.
  10. Lee, H., & Ma, J. (2022). "Mine Identification in GPR Images Using CNN." Journal of the Korean Society of Industrial and Applied Mathematics, Vol. 23, No. 6, pp. 82-88.
  11. Park, J. (2022). "A Study on the Development Direction of the Conscription System in Korea due to the Decrease in Military Resources." Korean Journal of Border Area Unification Studies, 126pg.
  12. Kim, H. (2023). "Analysis Study on the Operating and Maintenance Costs of the Mixed Military Service System as a Transition Direction for the Conscription System." Korean Journal of Institutional Economics, Vol. 2, Issue 48, pp. 35-61.
  13. Wikipedia. "Types, Specifications, and Actual Images of Mines.
  14. Military World of Yoo Yong-won. "Mine Detector-II Capable of Detecting Non-metallic Mines, including Wooden Mines / Hanwha Systems." Defense Acquisition Program Administration, 2020.
  15. Defense Acquisition Program Administration. "Non-metallic Mine Detection, Start of Mass Production of Mine Detector II, Image." 2021.
  16. Kim, M (2019), "Bark Identification Using a Deep Learning Model", Korea Multimedia Society, Volume 22 Issue 10, Pages.1133-1141 https://doi.org/10.9717/KMMS.2019.22.10.1133