6.Image denoising using scale mixtures of gaussians in the wavelet domain
J.Portila,V.strela,J.wainwright and E.P Simoncelli, IEEE transactions on Image Processing,vol.12,no.11,nov 2003
The image is corrupted by simulated additive white gaussian noise.The noisy image is then decomposed into subbands using steerable pyramid variant at different scales and orientations, each subband is then denoised by using gaussian scale mixtures, the parameters are estimated using maximum likelihood and bayes estimator. The low amplitude values are suppressed and retains the high amplitude values. Finally the denoised image is reconstructed.
This method provides better preservation of edges and other details even though it provides few artifacts.
Software: http://www.csee.wvu.edu/~xinl/source.html
This method provides better preservation of edges and other details even though it provides few artifacts.
Software: http://www.csee.wvu.edu/~xinl/source.html