SAR Image Filtering in Wavelet Domain by Subband Depended Shrink

This paper proposes an adaptive threshold estimation method for denoising in wavelet domain merged with translation invariant denoising. The subband shrink is computationally more efficient and adaptive because the parameters required for estimating the threshold depend on subband data. A new probability density function is proposed to model the statistics of wavelet coefficients. The subband threshold is derived using Bayesian estimation theory and the new pdf. Different shifts are used and applied to the noisy image in order to attain different estimates to the unknown image an then linearly average the estimates. Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. We apply the proposed method for speckle SAR images by using logarithmic transformation. Experimental results on several test images are compared with various denoising techniques.