한국수자원학회논문집 (Journal of Korea Water Resources Association)

  • 제54권12호
  • /
  • Pages.1233-1242
  • /
  • 2021
  • /
  • 2799-8746(pISSN)
  • /
  • 2799-8754(eISSN)
한국수자원학회 (Korea Water Resources Association)
권호 표지
DOI QR코드

DOI QR Code

비정형 데이터와 딥러닝을 활용한 내수침수 탐지기술 개발

Development of a method for urban flooding detection using unstructured data and deep learing

  • 이하늘 (인하대학교 스마트시티공학과) ;
  • 김형수 (인하대학교 사회인프라공학과) ;
  • 김수전 (인하대학교 사회인프라공학과) ;
  • 김동현 (인하대학교 스마트시티공학과) ;
  • 김종성 (인하대학교 스마트시티공학과)
  • Lee, Haneul (Program in Smart City Engineering, Inha University) ;
  • Kim, Hung Soo (Department of Civil Engineering, Inha University) ;
  • Kim, Soojun (Department of Civil Engineering, Inha University) ;
  • Kim, Donghyun (Program in Smart City Engineering, Inha University) ;
  • Kim, Jongsung (Program in Smart City Engineering, Inha University)
  • 투고 : 2021.09.23
  • 심사 : 2021.10.27
  • 발행 : 2021.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구에서는 비정형 데이터인 사진자료를 이용하여 침수의 발생여부를 판단하는 모델을 개발하였다. 침수분류를 모델 개발을 위하여 CNN기반의 VGG16, VGG19을 이용하였다. 모델을 개발하기 위하여 침수사진과 침수가 발생하지 않은 사진을 웹크롤링 방법을 이용하여 사진을 수집하였다. 웹크롤링 방법을 이용하여 수집한 데이터는 노이즈 데이터가 포함되어 있기 때문에 1차적으로 본 연구와 상관없는 데이터는 소거하였으며, 2차적으로 모델 적용을 위하여 224 × 224로 사진 사이즈를 일괄 변경하였다. 또한 사진의 다양성을 위해서 사진의 각도를 변환하여 이미지 증식을 수행하였으며. 최종적으로 침수사진 2,500장과 침수가 발생하지 않은 사진 2,500장을 이용하여 학습을 수행하였다. 모델 평가결과 모델의 평균 분류성능은 97%로 나타났으며. 향후 본 연구결과를 통하여 개발된 모델을 CCTV관제센터 시스템에 탑재한다면 신속하게 침수피해에 대한 대처가 이루어 질 수 있을 것이라 판단된다.

In this study, a model was developed to determine whether flooding occurred using image data, which is unstructured data. CNN-based VGG16 and VGG19 were used to develop the flood classification model. In order to develop a model, images of flooded and non-flooded images were collected using web crawling method. Since the data collected using the web crawling method contains noise data, data irrelevant to this study was primarily deleted, and secondly, the image size was changed to 224×224 for model application. In addition, image augmentation was performed by changing the angle of the image for diversity of image. Finally, learning was performed using 2,500 images of flooding and 2,500 images of non-flooding. As a result of model evaluation, the average classification performance of the model was found to be 97%. In the future, if the model developed through the results of this study is mounted on the CCTV control center system, it is judged that the respons against flood damage can be done quickly.

키워드

과제정보

이 논문은 행정안전부 재난피해 복구역량강화 기술개발사업의 지원을 받아 수행된 연구임(2021-MOIS36-002).

참고문헌

  1. Baydargil, H.B., Serdaroglu, S., Park, J.S., Park, K.H., and Shin, H.S. (2018). "Flood detection and control using deep convolutional encoder-decoder architecture." 2018 International Conference on Information and Communication Technology Robotics (ICT-ROBOT), IEEE, pp. 1-3.
  2. Bhadra, T., Chouhan, A., Chutia, D., Bhowmick, A., and Raju, P.L.N. (2020). "Flood detection using multispectral images and SAR data." International Conference on Machine Learning, Image Processing, Network Security and Data Sciences, Springer, Singapore, pp. 294-303.
  3. Choi, J., Park, K., Choi, S., and Jun, H. (2018). "A forecasting and alarm system for reducing damage from inland inundation in coastal urban areas: A case study of Yeosu City." Journal of the Korean Society of Hazard Mitigation, Vol. 18, No. 7, pp. 475-484. https://doi.org/10.9798/KOSHAM.2018.18.7.475
  4. Dias, D., and Dias, U. (2018). "Flood detection from social multimedia and satellite images using ensemble and transfer learning with CNN architectures." Proceedings of the MediaEval 2018 Workshop, CEUR, France, Vol. 2283, pp. 3-5.
  5. Godoy, D. (2018) Understanding binary cross-entropy/log loss: A visual explanation, accessed 13 September 2021,
  6. Hwang, S., Kim, K., Lee, G., and Lee, M. (2016). "A study on the development of automated damage estimation system using high resolution satellite imagery." Journal of Korean Society of Hazard Mitigation, Vol. 16, No. 2, pp. 161-172. https://doi.org/10.9798/KOSHAM.2016.16.2.161
  7. Jain, P., Schoen-Phelan, B., and Ross, R. (2020). "Automatic flood detection in SentineI-2 images using deep convolutional neural networks." Proceedings of the 35th Annual ACM Symposium on Applied Computing, ACM, NY, U.S., pp. 617-623.
  8. Jang, Y., and Chung, D. (2019). "Technology trend for image analysis based on deep learning." Current Industrial and Technological Trends in Aerospace, Vol. 17, No. 1, pp. 113-122.
  9. Jeon, H., Kim, J., Kim, K., and Hong, I. (2016). "Application of the LISFLOOD-FP model for flood stage prediction on the lower Mankyung River." Journal of Korea Water Resources Association, Vol. 49, No. 6, pp. 459-467. https://doi.org/10.3741/JKWRA.2016.49.6.459
  10. Kang, J., and Kwak, S. (2014). "Loitering detection solution for CCTV security system." Journal of Korea Multimedia Society, Vol. 17, No. 1, pp. 15-25. https://doi.org/10.9717/KMMS.2014.17.1.015
  11. Kim, S., Lee, S., Kim, T.W., and Kim, D. (2019). "Estimation of flooded area using satellite imagery and DSM terrain data." Journal of the Korean Society of Hazard Mitigation, Vol. 19, No. 7, pp. 471-483. https://doi.org/10.9798/kosham.2019.19.7.471
  12. Le, X.H., Ho, H.V., Lee, G., and Jung, S. (2019). "Application of long short-term memory (LSTM) neural network for flood forecasting." Water, Vol. 11, No. 7, 1387. https://doi.org/10.3390/w11071387
  13. LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). "Gradient-based learning applied to document recognition." Proceedings of the IEEE, Vol. 86, No. 11, pp. 2278-2324. https://doi.org/10.1109/5.726791
  14. Lee, B.J. (2017). "Analysis on inundation characteristics for flood impact forecasting in Gangnam drainage basin." Atmosphere, Vol. 27, No. 2, pp. 189-197. https://doi.org/10.14191/Atmos.2017.27.2.189
  15. Lee, J., Kang, B., Lee, H., Chung, Y., and Park, D. (2011). "A suspected criminal face identification via SVDD and SRC in surveillance system." Journal of KIISE: Computing Practices and Letters, Vol. 17, No. 2, pp. 135-139.
  16. Lee, S.H., Kang, D.H., and Kim, B.S. (2018). "A study on the method of calculating the threshold rainfall for rainfall impact forecasting." Journal of the Korean Society of Hazard Mitigation, Vol. 18, No. 7, pp. 93-102. https://doi.org/10.9798/kosham.2018.18.7.93
  17. Mason, D.C., Speck, R., Devereux, B., Schumann, G.J.P., Neal, J.C., and Bates, P.D. (2009). "Flood detection in urban areas using TerraSAR-X." IEEE Transactions on Geoscience and Remote Sensing, Vol. 48, No. 2, pp. 882-894. https://doi.org/10.1109/TGRS.2009.2029236
  18. Oga, T., Harakawa, R., Minewaki, S., Umeki, Y., Matsuda, Y., and Iwahashi, M. (2020). "River state classification combining patch-based processing and CNN." Plos One, Vol. 15, No. 12, e0243073. https://doi.org/10.1371/journal.pone.0243073
  19. Park, H.J. (2013). "A study on monitoring system for an abnormal behaviors by object's tracking." Journal of Digital Contents Society, Vol. 14, No. 4, pp. 589-596. https://doi.org/10.9728/DCS.2013.14.4.589
  20. Park, J., Dao, D.A., Kim, S., and Kim, D. (2019). "Detection of flood areas using Sentinel-1 satellite imagery and evaluation of its applicability." Journal of the Korean Society of Hazard Mitigation, Vol. 19, No. 6, pp. 53-61. https://doi.org/10.9798/kosham.2019.19.6.53
  21. Park, S.H., and Kim, H.J. (2020). "Design of artificial intelligence water level prediction system for prediction of river flood." Journal of the Korea Institute of Information and Communication Engineering, Vol. 24, No. 2, pp. 198-203. https://doi.org/10.6109/JKIICE.2020.24.2.198
  22. Ruder, S. (2016). "An overview of gradient descent optimization algorithms." arXiv preprint arXiv:1609.04747.
  23. Shin, D.H., Baek, J.W., Park, R.C., and Chung, K. (2021). "Deep learning-based vehicle anomaly detection using road CCTV data." Journal of the Korea Convergence Society, Vol. 12, No. 2, pp. 1-6. https://doi.org/10.15207/JKCS.2021.12.2.001
  24. Shin, H.J., Chae, H.S., Hwang, E.H., and Park, J.Y. (2012). "A study of informationization technique for detecting flood inundation area using RS." Journal of the Korean Association of Geographic Information Studies, Vol. 15, No. 1, pp. 172-183. https://doi.org/10.11108/KAGIS.2012.15.1.172
  25. Simonyan, K., and Zisserman, A. (2014). "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556.
  26. Singh, M.S.A.D.J., and Varnica, B. (2014). "Web crawler: Extracting the web data." International Journal of Computer Trends and Technology, Vol. 13, No. 3, pp. 132-137. https://doi.org/10.14445/22312803/IJCTT-V13P128
  27. Song, Y.H., Song, Y.S., Park, M.J., and Lee, J.H. (2014). "Flood forecasting estimation methodology of standard rainfall for urban mid and small rivers considering upper-and down-stream water levels." Journal of the Korean Society of Hazard Mitigation, Vol. 14, No. 2, pp. 289-298. https://doi.org/10.9798/KOSHAM.2014.14.2.289
  28. Theera-Umpon, N., Auephanwiriyakul, S., Suteepohnwiroj, S., Pahasha, J., and Wantanajittikul, K. (2008). "River basin flood prediction using support vector machines." 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), IEEE, Hong Kong, China, pp. 3039-3043.
  29. Tsakiri, K., Marsellos, A., and Kapetanakis, S. (2018). "Artificial neural network and multiple linear regression for flood prediction in Mohawk River." Water, Vol. 10, No. 9, 1158. https://doi.org/10.3390/w10091158
  30. Wu, C.L., and Chau, K.W. (2006). "A flood forecasting neural network model with genetic algorithm." International Journal of Environment and Pollution, Vol. 28, No. 3-4, pp. 261-273. https://doi.org/10.1504/IJEP.2006.011211