A New Adaptive Algorithm for Despeckling SAR Images Based on Contourlet Transform

A new adaptive contourlet transform-based technique for SAR image speckles removal is presented. It relies on realistic distribution of the SAR image's contourlet coefficients that represent mainly speckle noise on one hand and those that represent the useful signals corrupted on the other hand. The analytic model for these distributions is proposed. The shrinkage factor for despeckling is estimated at neighbourhood of the reference contourlet coefficient in each sub bands. Finally, a comparison of performance of Lee filter, Forster filter, Gamma filter, wavelet-based despeckling and contourlet transform-based despeckling is provided for both simulated and actual SAR images. It shows that the contourlet methods strongly suppress speckle, while preserving image details and sharpness