The application of wavelet transform for speckle suppression in radar imagery
Y. Dong, Bruce C. Forster, A. Milne, Catherine Ticehurst · 2002
An effective algorithm for speckle noise smoothing using wavelet transform techniques is presented. Recursive wavelet transforms are used to gradually suppress speckle noise. It is found that this algorithm is more powerful compared to other existing filtering algorithms in terms of speckle suppression for synthetic aperture radar images. Examples show that the ratio of the original standard deviation to mean of about 0.30 (equivalent to 3-look images) can be reduced to 0.05-0.03 (equivalent to more than 100-look images), with a possible small sacrifice of losing some details and narrow edges.