A Structure Based Speckle Filtering Method for Ocean SAR Images
Pu Cheng, Zhentao Yu, Jie Chen · 2021
Synthetic aperture radar (SAR) images are subjected to speckle noise. Ocean SAR imagery contains complex and delicate features. It is much more difficult to suppress the speckle in some case. Efficient speckle reduction requires a big homogeneous area to average the pixels. At the same time, the subtle features have to be maintained. Here we propose a simple but efficient method which incorporates the minimum mean-square error (MMSE) estimator and the non-local mean method. The speckle can be suppressed much more efficiently as more homogeneous pixels can be found by using the non-local mean method. The subtle structures can be preserved by using the MMSE estimator.