SAR Images Despeckling via Bayesian Shrinkage Based on Nonsubsampled Contourlet Transform

We propose a novel and efficient SAR image despeckling via Bayesian shrinkage based on nonsubsampled contourlet transform, which has been recently introduced. Despeckling by means of contourlet transform introduce many visual artifacts due to the Gibbs-like phenomena. Nonsubsampled contour let transform is a flexible multiscale, multidirection and shift-invariant image decomposition that can be efficiently implemented via transform. A Bayesian estimator is applied to the decomposed contour let coefficients of the logarithmically transformed image to estimate the best value for the noise-free signal. Experimental results show that compared with conventional wavelet despeckling algorithm, the proposed algorithm can achieve an excellent balance between suppresses speckle effectively and preserves image details, and the significant information of original image like textures and contour details is well maintained.