SAR Images Denoising Based on Rough Set Theory in Contourlet Domain

The paper present a speckle reduction method based on rough set theory in contourlet domain. The modified upper approximation set is proposed to check good continuation of the window center with a rotating neighborhood templates distributed uniformly around the center. Firstly a logarithmically transform is applied for converting the speckles to additive noises, then the contourlet transform is employed for decomposing the logarithmically transformed image into low frequency samples and directional bandpass samples. A non-statistical variance between the center of a window and pixels of window template is used as the measure of good continuation for bandpass samples in each scale. The minimum vale of this variance is found to get the most homogeneous template and the mean of this most homogeneous template is then used to take place the center coefficient. Finally the de-noised image is reconstructed by the inverse contourlet transform and the experiment result shows this method is effectively for SAR images de-noising.