DOI QR코드

DOI QR Code

A Novel Water Surface Detection Method Based on Correlation Analysis for Rectangular Control Area

직사각형 검사영역의 상관도 분석을 통한 수면위치 탐색 방법

  • Lee, Chan Joo (River and Coastal Research Division, Korea Institute of Construction Technology) ;
  • Seo, Myoung Bae (ICT Convergence and Integration Research Division, Korea Institute of Construction Technology) ;
  • Kim, Dong Gu (River and Coastal Research Division, Korea Institute of Construction Technology) ;
  • Kwon, Sung Il (River and Coastal Research Division, Korea Institute of Construction Technology)
  • 이찬주 (한국건설기술연구원 하천해안연구실) ;
  • 서명배 (한국건설기술연구원 ICT융합연구실) ;
  • 김동구 (한국건설기술연구원 하천해안연구실) ;
  • 권성일 (한국건설기술연구원 하천해안연구실)
  • Received : 2012.06.12
  • Accepted : 2012.08.17
  • Published : 2012.12.31

Abstract

In this study, a novel water surface detection method was proposed. In the method water surface is detected by analysis on correlation coefficients obtained from rectangular control areas of the same vertical position in two successive images including both water surface and staff gauge. Four methods respectively based on threshold, peak, slope and variance ratio, are used to identify water surface from vertical distribution of correlation coefficient. In addition, swaying correction algorithm and statistical filtering are applied to minimize outliers caused by positional image mismatch. Images taken from 28 different sites during low flow were tested to evaluate the method. Mean relative error to eye measurement was approximately from 3.4 to 5.7 cm. As long as water surface moves, this method can be used to improve image stage gauge by supplementing the previous water surface detection method.

본 연구에서는 목자판과 수면이 포함되어 있는 시차를 가진 두 영상에 대해 직사각형 검사영역을 설정하고 그 상관계수를 분석하여 수면을 판단하는 새로운 수면인식 기법을 제안하였다. 상관계수의 수직적인 값들로부터 임계치, 첨두값, 기울기, 분산비 등 4가지 방법을 이용하여 수면의 위치를 판정하였다. 흔들림 등으로 인해 두 영상의 위치가 불일치하여 생기는 문제를 제거하기 위해 추가로 영상의 흔들림을 보정하는 알고리즘과 통계적 필터링 기법을 적용하였다. 저수시에 촬영한 28개 지점의 영상에 개발한 수면 인식 방법을 적용하였다. 이 방법으로 계산한 수면은 목측 수면과의 평균상대오차가 3.4~5.7 cm 정도로 나타났다. 수면의 요동이 있을 경우, 이 방법은 기존 방법을 보완하여 영상수위계의 수위 측정성능을 높이는데 활용될 수 있을 것이다.

Keywords

References

  1. Dinehart, R.L., and Burau, J.R. (2005). "Repeated surveys by acoustic Doppler current profiler for flow and sediment dynamics in a tidal river." Journal of Hydrology, Vol. 314, pp. 1-21. https://doi.org/10.1016/j.jhydrol.2005.03.019
  2. FRANCE ETAT PONTS CHAUSSEES (2005). Water level measurement device for e.g. river, has storage and processing circuit processing image taken by camera to provide information representative of liquid level with respect to reference level, France Patent (FR 2865802A1).
  3. Katsanos, M. (2008). Intermarket Trading Strategies, John Wiley and Sons, Chichester, p. 412.
  4. Kim, J.D., Han, Y.J., and Hahn, H.S. (2010) "Image-based Water Level Measurement Method Adapting to Ruler's Surface Condition." Journal of The Korea Society of Computer and Information, Vol. 15, No. 9. pp. 67-76 (in Korean). https://doi.org/10.9708/jksci.2010.15.9.067
  5. Kim, W., Kim, C.Y., Kim, D.G., and Lee, C.J. (2006) "Development of the Image Stage Gauge using the Image Process Technique." Proceedings of the Korea Water Resources Association Conference 2006, pp. 500-504 (in Korean).
  6. Kim, W., Kim, C.Y., Lee, C.J., and Kim, D.G. (2007) "Practical Application of Image Stage Gauge." Proceedings of the Korea Water Resources Association Conference 2007, pp. 560-564 (in Korean).
  7. Kim, W., Kwon, S.I., Kim, D.G., Lee, C.J., Kim, Y.J., and Maeng, E.A. (2011). Development and Operation of Image Stage Gauge, Technical Report, TR 2011-13, SWRRC (in Korean).
  8. Kwon, S.I., Kim, W., Lee, C.J., and Kim, S.D. (2010) "Water Level Measurement Method Based on Temporal Variation of Water Surface Pixel Arrangement in Successive Images." Journal of Korea Water Resources Association, Vol. 43, No. 9. pp. 791-787 (in Korean). https://doi.org/10.3741/JKWRA.2010.43.9.781
  9. Kwon, S.I., Lee, C.J., Kim, D.G., and Seo, M.B. (2012) "A Technique for Water Surface Level Detection by Analyzing Motion in Water Area." Proceedings of the Korea Water Resources Association Conference 2012 (in Korean).
  10. Otsu, N. (1979). "A threshold selection method from gray-level histograms." IEEE Trans. Sys., Man., Cyber. Vol. 9, No. 1, pp. 62-66. https://doi.org/10.1109/TSMC.1979.4310076
  11. Ryou, K.K., Yoon, B.M., and Jeong, B.S. (2008) "A Surface Image Velocimetry Algorithm for Analyzing Swaying Images." Journal of Korea Water Resources Association, Vol. 41, No. 8. pp. 855-862 (in Korean). https://doi.org/10.3741/JKWRA.2008.41.8.855
  12. Saito, A., Iwahashi, M. (2006). Water level detection algorithm based on synchronous frame addition and filtering, Proceedings of the 19th workshop on circuits and systems in Karuizawa, pp. 525-530 (in Japanese).
  13. Seo, M.B., Lee, C.J., Kwon, S.I., and Kim, D.G. (2012) "An Application and Evaluation of a Water Level Detection Method by Correlation Analysis for Successive Watermark Images." Proceedings of Korea Computer Congress 2012, Vol. 39, No. 1(B), pp. 43-45 (in Korean).
  14. Takagi, Y., Tsujikawa, A., Takato, M., Saito, T., and Kaida, M. (1998). "Development of a noncontact liquid level measurement system using image processing." Water Science and Technology, Vol. 37, No. 12, pp. 381-387. https://doi.org/10.1016/S0273-1223(98)00356-4
  15. Takagi, Y., Yoneoka, T., Mori, H., Tujikawa, A., Saito, T., and Karube, K. (2000). "Research on dam water level measurement technology by means of a visual sensor." The Society of Environmental Instrumentation, Control and Automation, Vol. 5, No. 2, pp 179- 188 (in Japanese).

Cited by

  1. Measuring Inundation Depth in a Subway Station Using the Laser Image Analysis Method vol.10, pp.11, 2018, https://doi.org/10.3390/w10111558