Refinement to a wavelet-based algorithm for the estimation of velocity vector fields
Document Type
Conference Proceeding
Publication Date
12-1-2004
Abstract
This paper presents a refinement to the algorithm that estimates velocity vectors fields from sequence of images. This algorithm was introduced in a previous publication of these series [8]. The algorithm uses wavelet analysis to decompose sea surface temperature satellite images into complementary sub-images of decreasing resolution. Maximization of cross correlation is used to detect similarities between decomposed sub-images as indicators for likely displacements of ocean features. The algorithm performed well under controlled circumstances. However, when tested with SST images from the Gulf of Mexico some discrepancies arose between the algorithm estimations and corresponding drifter data. In this paper we discuss some of these discrepancies and introduce the idea of reference vector fields to improve estimations.
Publication Source (Journal or Book title)
Proceedings of the International Conference on Imaging Science Systems and Technology CISST 04
First Page
363
Last Page
369
Recommended Citation
Tonsmann, G., Walker, N., Rouse, L., Tyler, J., & Wiseman, W. (2004). Refinement to a wavelet-based algorithm for the estimation of velocity vector fields. Proceedings of the International Conference on Imaging Science Systems and Technology CISST 04, 363-369. Retrieved from https://repository.lsu.edu/oceanography_coastal_pubs/1301