An adaptive method for estimation of wave height based on statistics of sea surface images

The significant wave height H_mo can be estimated by radar image statistics of sea surface images, based on texture analysis. 17 different parameters calculated from image statistics have been considered, and it is shown that number of extrema per area is the parameter giving the best estimates in situations with no heavy precipitation. The estimates, however, become inaccurate in situations with images cluttered by heavy precipitation. This article describes a method for improving the estimation of H_mo. A method for detecting the problematic situations is developed, and then parameters being able to make a better estimate in the same situations are evaluated