한국수자원학회논문집 (Journal of Korea Water Resources Association)

  • 제56권4호
  • /
  • Pages.245-259
  • /
  • 2023
  • /
  • 2799-8746(pISSN)
  • /
  • 2799-8754(eISSN)
한국수자원학회 (Korea Water Resources Association)
권호 표지
DOI QR코드

DOI QR Code

농업용 저수지 CCTV 영상자료 기반 수위 인식 모델 적용성 검토

A study on the application of the agricultural reservoir water level recognition model using CCTV image data

  • 권순호 (고려대학교 4단계 BK21 건축사회환경공학교육연구단) ;
  • 하창용 (한국농어촌공사 경북지역본부) ;
  • 이승엽 (한남대학교 공과대학 토목환경공학과)
  • Kwon, Soon Ho (Future-Shapers for Solving Global Problems in Construction, Korea University) ;
  • Ha, Changyong (Gyeongbuk Regional Headquarter, Korea Rural Community Corporation) ;
  • Lee, Seungyub (Department of Civil and Environmental Engineering, Hannam University)
  • 투고 : 2023.01.31
  • 심사 : 2023.03.02
  • 발행 : 2023.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

농업용 저수지는 농업용수 공급에 있어서 매우 중요한 생산기반시설로, 우리나라 농업용수의 60% 정도를 공급하고 있다. 다만, 여러 문제로 인해 농업용수의 효율적인 공급에 어려움이 발생하고 있으며, 효과적인 공급 및 관리 체계 구현을 위한 정확한 실시간 저수위 혹은 저수량 추정이 필요하다. 본 연구에서는 영상정보를 활용한 딥러닝 기반 농업용 저수지 수위 인식 모델을 제안하였다. 개발한 모델은 (1) CCTV 영상정보 자료 수집 및 분석, (2) U-Net 이미지 분할 방법을 통한 입력 자료 생성, 그리고 (3) CNN과 ResNet 모델을 통한 수위 인식 세 단계로 구성된다. 모델은 두 농업용 저수지(G저수지와 M저수지)의 영상자료와 저수위 시계열자료를 활용하여 구현하였다. 적용 결과 이미지 분할 모델의 성능은 매우 우수한 것으로 나타났으며, 수위 인식 모델의 경우 수위 분류 계급구간에 따라 성능이 상이한 것으로 나타났다. 특히 영상자료의 픽셀 변동이 클수록 정확도 80% 이상이 확보 가능한 것으로 확인되었으나, 그렇지 않은 경우, 정확도가 50% 수준인 것으로 나타났다. 본 연구에서 개발한 모델은 향후 이미지 자료가 추가로 확보될 경우, 그 활용도 및 정확도가 더 높아질 것으로 기대한다.

The agricultural reservoir is a critical water supply system in South Korea, providing approximately 60% of the agricultural water demand. However, the reservoir faces several issues that jeopardize its efficient operation and management. To address this issues, we propose a novel deep-learning-based water level recognition model that uses CCTV image data to accurately estimate water levels in agricultural reservoirs. The model consists of three main parts: (1) dataset construction, (2) image segmentation using the U-Net algorithm, and (3) CCTV-based water level recognition using either CNN or ResNet. The model has been applied to two reservoirs G-reservoir and M-reservoir with observed CCTV image and water level time series data. The results show that the performance of the image segmentation model is superior, while the performance of the water level recognition model varies from 50 to 80% depending on water level classification criteria (i.e., classification guideline) and complexity of image data (i.e., variability of the image pixels). The performance of the model can be improved if more numbers of data can be collected.

키워드

과제정보

이 성과는 2021년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(No. 2021R1C1C2004896).

참고문헌

  1. Badrinarayanan, V., Kendall, A., and Cipolla, R. (2017). "Segnet: A deep convolutional encoder-decoder architecture for image segmentation." IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, No. 12, pp. 2481-2495. https://doi.org/10.1109/TPAMI.2016.2644615
  2. Bang, J., Choi, J.-Y., Yoon, P., Oh, C.-J., Maeng, S.-J., Bae, S.-J., Jang, M.-W., Jang, T., and Park, M.S. (2021). "Assessing irrigation water supply from agricultural reservoir using automatic water level data of irrigation canal." Journal of The Korean Society of Agricultural Engineers, Vol. 63, No. 1, pp. 27-35.
  3. Chaudhary, P., D'Aronco, S., Moy de Virty, M., Leitao, J.P., and Wegner, J.D. (2019). "Flood-water level estimation from social media images." The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, IV-2/W5, pp. 5-12. doi: 10.5194.isprs-annals-IV-2-W5-5-2019. https://doi.org/10.5194.isprs-annals-IV-2-W5-5-2019
  4. Chen, L.C., Papandreou, G., Schroff, F., and Adam, H. (2017). "Rethinking atrous convolution for semantic image segmentation." arXiv Preprint arXiv,1706.05587. doi: 10.48550/arXiv.1706.05587.
  5. Creutin, J.D., Muste, M., Bradley, A.A., Kim, S.C., and Kruger, A. (2003). "River gauging using PIV techniques: A proof of concept experiment on the Iowa River." Journal of Hydrology, Vol. 277, No. 3-4, pp. 182-194. https://doi.org/10.1016/S0022-1694(03)00081-7
  6. Dal Sasso, S. F., Ljubicic, R., Pizarro, A., Pearce, S., Maddock, I., and Manfreda, S. (2023). "Image-based velocity estimations under different seeded and unseeded river flows." EGU General Assembly 2023, EGU, Vienna, Austria. doi: 10.5194/egusphereegu23-6936, 2023.
  7. Dhara, S., Dang, T., Parial, K., and Lu, X.X. (2020). "Accounting for uncertainty and reconstruction of flooding patterns based on multi-satellite imagery and support vector machine technique: A case study of Can Tho City, Vietnam." Water, Vol. 12, No. 6, 1543.
  8. He, K., Zhang, X., Ren, S., and Sun, J. (2016). "Deep residual learning for image recognition." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Las Vegas, NV, U.S.,pp. 770-778.
  9. Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R.R. (2012). "Improving neural networks by preventing co-adaptation of feature detectors." arXiv Preprint arXiv,1207.0580.
  10. Hong, E.M., Nam, W.H., Choi, J.Y., and Kim, J.T. (2014). "Evaluation of water supply adequacy using real-time water level monitoring system in paddy irrigation canals." Journal of the Korean Society of Agricultural Engineers, Vol. 56, No. 4, pp. 1-8. https://doi.org/10.5389/KSAE.2014.56.4.001
  11. Jenks, G.F. (1967). "The data model concept in statistical mapping." International Yearbook of Cartography, Vol. 7, pp. 186-190.
  12. Kim, K.H., Kim, M.G., Yoon, P.R., Bang, J.H., Myoung, W.H., Choi, J.Y., and Choi, G.H. (2022). "Application of CCTV image and semantic segmentation model for water level estimation of irrigation channel." Journal of The Korean Society of Agricultural Engineers, Vol. 64, No. 3, pp. 63-73. https://doi.org/10.5389/KSAE.2022.64.3.063
  13. Kim, S.J., Kwon, H.J., Kim, J., and Kim, P.S. (2016). "Economical design of water level monitoring network for agricultural water quantification." Journal of The Korean Society of Agricultural Engineers, Vol. 58, No. 5, pp. 19-28.
  14. Kingma, D.P., and Ba, J. (2014). "Adam: A method for stochastic optimization." arXiv Preprint arXiv, 1412.6980.
  15. Le Boursicaud, R., Penard, L., Hauet, A., Thollet, F., and Le Coz, J. (2016). "Gauging extreme floods on YouTube: Application of LSPIV to home movies for the post-event determination of stream discharges." Hydrological Processes, Vol. 30, No. 1, pp. 90-105. https://doi.org/10.1002/hyp.10532
  16. LeCun, Y., and Bengio, Y. (1995). "Convolutional networks for images, speech, and time series." The Handbook of Brain Theory and Neural Networks, Vol. 3361, No. 10, 1995.
  17. Lee, J., Noh, J., Kang, M., and Shin, H. (2020). "Evaluation of the irrigation water supply of agricultural reservoir based on measurement information from irrigation canal." Journal of The Korean Society of Agricultural Engineers, Vol. 62, No. 6, pp. 63-72.
  18. Li, W., Liao, Q., and Ran, Q. (2019). "Stereo-imaging LSPIV (SILSPIV) for 3D water surface reconstruction and discharge measurement in mountain river flows." Journal of Hydrology, Vol. 578, 124099.
  19. Li, Z., Lu, C.Z., Qin, J., Guo, C.L., and Cheng, M.M. (2022). "Towards an end-to-end framework for flow-guided video inpainting." In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, New Orleans, LA, U.S., pp. 17562-17571.
  20. Long, J., Shelhamer, E., and Darrell, T. (2015). "Fully convolutional networks for semantic segmentation." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Boston, MA, U.S., pp. 3431-3440.
  21. Maehara, H., Nagase, M., and Taira, K. (2016). "Water level measurement from CCTV camera images using water gauge images taken at the time of low water level." Journal of the Japan Society of Photogrammetry and Remote Sensing, Vol. 55, No. 1, pp. 66-68 (in Japanese). doi: 10.4287/jsprs.55.66.
  22. Maehara, H., Nagase, M., Kuchi, M., Suzuki, T., and Taira, K. (2019). "A deep-learning based water-level measurement method from CCTV camera images." Journal of the Japan Society of Photogrammetry and Remote Sensing, Vol. 58, No. 1, pp. 28-33 (in Japanese). doi: 10.4287/jsprs.58.28.
  23. Ministry of Environment (ME) (2020) Korean climate change assessment report 2020 - Scientific evidence for climate change -. Korea Meteorological Administration.
  24. Ministry of Environment (ME) (2022) Drought information analysis annual report.
  25. Ministry of Land, Transport and Maritime Affairs (MLTMA) (2011) Improvement and supplement research of probability rainfall map.
  26. Muste, M., Ho, H.C., and Kim, D. (2011). "Considerations on direct stream flow measurements using video imagery: Outlook and research needs." Journal of Hydro-environment Research, Vol. 5, No. 4, pp. 289-300. https://doi.org/10.1016/j.jher.2010.11.002
  27. Perks, M.T., Dal Sasso, S.F., Hauet, A., Jamieson, E., Le Coz, J., Pearce, S., Pena-Haro, S., Pizarro, A., Strelnikova, D., Tauro, F. and Bomhof, J. (2020). "Towards harmonisation of image velocimetry techniques for river surface velocity observations." Earth System Science Data, Vol. 12, No. 3, pp. 1545-1559. https://doi.org/10.5194/essd-12-1545-2020
  28. Pumo, D., Alongi, F., Ciraolo, G., and Noto, L.V. (2021). "Optical methods for river monitoring: A simulation-based approach to explore optimal experimental setup for LSPIV." Water, Vol. 13, No. 3, 247.
  29. Ronneberger, O., Fischer, P., and Brox, T. (2015). "U-net: Convolutional networks for biomedical image segmentation." In Proceedings of Medical Image Computing and Computer-Assisted Intervention, Springer International Publishing, Munich, Germany, pp. 234-241.
  30. Serte, S., Serener, A., and Al-Turjman, F. (2022). "Deep learning in medical imaging: A brief review." Transactions on Emerging Telecommunications Technologies, Vol. 33, No. 10, e4080.
  31. Simonyan, K., and Zisserman, A. (2014). "Very deep convolutional networks for large-scale image recognition." arXiv Preprint arXiv, 1409.1556.
  32. Szegedy, C., Reed, S., Erhan, D., Anguelov, D., and Ioffe, S. (2014). "Scalable, high-quality object detection." arXiv Preprint arXiv, 1412.1441.
  33. Tan, M., and Le, Q. (2019). "Efficientnet: Rethinking model scaling for convolutional neural networks." In International Conference on Machine Learning, PMLR, Long Beach, CA, U.S., pp. 6105-6114.
  34. Yang, M., Nam, W., Kim, H., Kim, T., Shin, A., Kang, M. (2021). "Anomaly detection in reservoir water level data using the LSTM model based on deep learning." Journal of the Korean Society of Hazard Mitigation, Vol. 21, No. 1, pp. 71-81. https://doi.org/10.9798/KOSHAM.2021.21.1.71