Image denoising using scale mixtures of Gaussians in the wavelet domain

Abstract

We describe a method for removing noise from digital images, based on a statistical model of the coefficients of an overcomplete multiscale oriented basis. Neighborhoods of coefficients at adjacent positions and scales are modeled as the product of two independent random variables: a Gaussian vector and a hidden positive scalar multiplier. The latter modulates the local variance of the coefficients in the neighborhood, and is thus able to account for the empirically observed correlation between the coefficient amplitudes. Under this model, the Bayesian least squares estimate of each coefficient reduces to a weighted average of the local linear estimates over all possible values of the hidden multiplier variable. We demonstrate through simulations with images contaminated by additive white Gaussian noise that the performance of this method substantially surpasses that of previously published methods, both visually and in terms of mean squared error.

DOI: 10.1109/TIP.2003.818640

Extracted Key Phrases

9 Figures and Tables

0100200'04'06'08'10'12'14'16
Citations per Year

1,900 Citations

Semantic Scholar estimates that this publication has 1,900 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{Portilla2003ImageDU, title={Image denoising using scale mixtures of Gaussians in the wavelet domain}, author={Javier Portilla and Vasily Strela and Martin J. Wainwright and Eero P. Simoncelli}, journal={IEEE transactions on image processing : a publication of the IEEE Signal Processing Society}, year={2003}, volume={12 11}, pages={1338-51} }