Spectrum-Correlation Method and Filters Technique Application for the Velocity Field Estimation
A new image processing procedure by using Spectrum-Correlation has been developed to estimate an instantaneous velocity distribution. This has been applied to analyze velocity fields under breaking waves. As in an image-correlation method, this procedure finds out similar patterns from two successive pictures and estimates velocity vectors. The used pictures were taken with very short time lag between them. Therefore the procedure does not require to trace fine particles, and can be applied under various conditions. The processing time for evaluating the velocity distribution with a personal computer is relatively short. To improve the accuracy of the evaluated velocity distribution, a Laplacian filter and a smoothing filter are introduced. The Laplacian filter emphasizes image patterns near the water surface, and the smoothing filter reduces noises due to the flow rotating motions and turbulence.
KeywordsVortex Crest Aliasing
Unable to display preview. Download preview PDF.
- 1.Nishimura, Hitoshi and Satoshi Takewaka (1988): Estimation of a velocity field in a breaking wave legion from VTR-image data using a correlation analysis (Japanese), Proc. of Coastal Eng., JSCE, No. 35, pp. 45–48.Google Scholar
- 2.Mano, Akira and Shigeya Kamio (1990): Flow field analysis on air-entrained region of breaking waves by correlation method (Japanese), Proc. of JSCE, No. 423/II-14, pp. 171–180.Google Scholar
- 3.Okuno, Taketoshi and Taminori Kinoshita (1987): Development of a velocity measurement method using image processing technique (Japanese), Assoc. for the Study of Flow Measurement, Osaka, 4, p. 177–84.Google Scholar
- 4.Hino, Mikio (1991): Some Improvements on the Velocity Estimation by FFT and Spectrum-Correlation Method, Tech. Rep., Dept. of Civil Eng., Tokyo Institute of Technology, No. 44.Google Scholar