Synthetic-Aperture Radar Image Despeckling based on Improved Non-Local Means and Non-Subsampled Shearlet Transform

When Synthetic- Aperture Radar (SAR) image is transformed into wavelet domain and other transform domains, most of the coefficients of the image are small or zero. This shows that SAR image is sparse. However, speckle can be seen in SAR images. The non-local means (NLM) is a despeckling algorithm, but it cannot overcome the speckle in homogeneous regions and it blurs edge details of the image. In order to solve these problems, an improved NLM is suggested by using L-1/2 norm instead of L-2 norm as the measure of similarity. At the same time, the non-subsampled Shearlet transform (NSST) is chosen for effective speckle suppression in edge regions. By combining NSST with improved NLM, a new type of despeckling algorithm is proposed. Experiments demonstrate that the new algorithm leads to satisfying results for SAR images.