在山區發生的土石流事件常常造成當地居民很大的生命及財產的損失，尤其是每逢到颱風季節或者忽然下起連日豪雨，就很可能會造成土石流的產生。為了防範這些巨大的天然災害，水土保持局在這些可能發生土石流的區域架設了大量的監視設備，以監控山區土石以及河道改變的情況。本篇論文的目的就是希望藉由這些監視設備所拍攝到的土石流影片去估算出土石流的流動速度，我們主要是使用數種向量分析(Motion Estimation)的方法先估算出運動向量(Motion Vector)，發現運動向量在有水流動的影片中會有方向一致性，用K平均演算法(K-mean algorithm)和其他過濾機制來過濾出真正有意義的運動向量，再從這些有意義的運動向量去估算出水流的速度。我們首先嘗試針對人造影像和自行拍攝的河川影像資料去做實驗，人造影像和在河川影像上所估算的流速和其真實值比較的結果相差不會太大，可由此證明本演算法可以對流水影像估算出準確度還算不錯的流速，進而在把此演算法套用在土石流影像上，把估算出的流速和用人眼觀測的流動快慢去做比較，其比較結果確實也是很接近的。

[1] Y. Itakura, K. Ogawa, H. Suwa, and K. Mizuhara, “Trends and fluctuations of the surface-velocity of the surface-velocity of debris flow measured by a non-contact speed sensor with a spatial filter,” Proceedings of the International Symposium on Fluid Control and Measurement, Pergamon Press, New York, pp. 781-786, 1986.
[2] Y. Itakura and H. Suwa, “Measurement of surface velocity of debris flow by spatial filtering velocimetry,” Proceedings of The Japan-China Symposium on Landslides and Debris Flows, (Japan Landslids Society, Tokyo), pp. 199-203, 1989.
[3] Y. Itakura, H. Inaba, and M. Kasahara, “Measurement of surface velocity of natural turbulent flows by spatial filtering velocimetry,” Proceedings of the XIII IMEKO World Congress, Budapest, Hungary, pp. 2086-2091, 1994.
[4] J. T. Ator, “Image-velocity sensing with parallel-slit reticles,” Journal of the Optical Society of America, vol. 53, pp.1416-1422, 1963.
[5] J. T. Ator, “Image velocity sensing by optical correlation,” Applied Optics, vol. 5, pp. 1325-1331, 1966.
[6] Y. Aizu and T. Asakura, “Principles and development of spatial filtering velocimetry,” Applied Physics B: Lasers and Optics, vol. 43, pp.209-224, 1987.
[7] Uddin M. S., Inaba H., Itakura Y., and Kasahara M., “Estimation of the surface velocity of debris flow with computer-based spatial filtering,” Applied Optics, vol. 37, pp. 6234-6239, 1998.
[8] M. S. Uddin, H. Inaba, Y. Itakura, Y. Yoshida, and M. Kasahara, “Adaptive computer-based spatial filtering method for more accurate estimation of the surface velocity of debris flow,” Applied Optics, vol. 38, pp. 6714-6721, 1999.
[9] B. K. P Horn and B. G.. Schunck, “Determining optical flow,” Artificial Intelligence, vol. 17, pp. 185-203, 1981.
[10] S. Ando, “A velocity vector field measurement system based on spatio-temporal image derivative,” Trans. Soc. Instrum. Control Eng ., vol. 22, no. 12, pp.1330-1336, 1986.
[11] J. K. Kearney, W. B. Thompson, and D. L. Boley, “Optical flow estimation: An error analysis of gradient-based methods with local optimization,” IEEE Trans. Pattern Anal. Machine Intell. ,vol. PAMI-9, no. 2, pp. 229-244, 1987.
[12] C. Stiller and J. Konrad, “Estimating motion in image sequences,” IEEE Signal Process. Magazine, vol. 16, pp. 70-91, July 1999.
[13] H. Inaba, M. S. Uddin, Y. Itakura, and M. Kasahara, “Surface velocity vector field measurement of debris flow based on spatio temporal derivative space method,” Proc. 1st International Conference on Debris-Flow Hazards Mitigation, California, USA, pp. 757-766, 1997.
[14] H. Inaba, Y. Itakura, and M. Kasahara, “Surface velocity computation of debris flow by vector field measurement,” Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere, vol. 25, pp. 741-744, 2000.
[15] M. S. Uddin, H. Inaba, Y. Yoshida, and Y. Itakura, “Notes on gradient-based methods for estimation of large motion in image sequences,” in Proceedings of the IEEE International Conference on Electrical and Computer Engineering, Dhaka, Bangladesh, pp. 74-77, January 2001.
[16] M. S. Uddin, H. Inaba, Y. Itakura, Y. Yoshida, and M. Kasahara, “Large motion estimation by gradient technique – application to debris flow velocity field,” Physics and Chemistry of the Earth, Part C: Solar, Terrestrial & Planetary Science, vol. 26, no. 9, pp.633-638, 2001.
[17] P. T. Tokumaru and P. E. Dimotakis, “Image correlation velocimetry,” Experiments in Fluids, vol. 19, pp. 1-15, 1995.
[18] A. K. Jain and J. R. Jain, “Displacement measurement and its application in interframe image coding,” IEEE Transactions on Communications, vol. 29, pp. 1799-1808, 1981.
[19] Yun Q. Shi and Huifang Sun, Image and Video Compression for Multimedia Engineering, CRC Press, New York, 2000, chap. 11.
[20] A. K. Jain, Fundamentals of Digital Image Processing , Prentice-Hall, Englewood Cliffs, N.J. 1989, pp. 400-407.
[21] M. S. Uddin, H. Inaba, Y. Yoshida, and Y. Itakura, “Debris flow velocity estimation: a comparison between gradient-based method and cross-correlation method,” Image Processing: Algorithms and Systems, vol. 4667, pp. 130-138, 2002.
[22] D. G. Childers, ed., Modern Spectrum Analysis, Institute of Electrical and Electronics Engineers, Piscataway, N.J. , 1978, pp. 23-118.
[23] H. Ogura and Y. Yoshida, “Spectral analysis and subtraction of noise in radar signals,” IEEE Transactions on Aerospace and Electronic Systems, vol. AES-17, pp. 62-71, 1981.
[24] Video Document of Experiments at the USGS Debris-Flow Flume Available from http://pubs.usgs.gov/of/2007/1315/.