SAR Image Despeckling via Bivariate Shrinkage Based on Directionlet Transform

Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. A novel and efficient SAR image despeckling algorithm based on Directionlet transform using bivariate shrinkage is proposed to remove speckle noise while preserving the structural features and textural information of the scene. First, an anisotropic directionlet transform is taken on the logarithmically transformed SAR images. The distribution of speckle noise is modeled as an additive Gaussian distribution with zero-mean. Then, a bivariate shrinkage with local variance estimation is applied to the decomposed directionlet coefficients of the logarithmically transformed image to estimate the best value for the noise-free signal. Finally, the performance of the proposed algorithm is compared with those of existing despeckling methods applied on both synthetic speckled images and actual SAR images. Experimental results show that compared with conventional wavelet and contourlet despeckling algorithm, the proposed algorithm can keep the better balance between suppresses speckle effectively and preserves image details, and the important feature of original image like textures and contour details is well maintained.