An elliptically contoured exponential mixture model for wavelet based image denoising

An elliptically contoured exponential distribution is developed as a generalization of the univariate Laplacian distribution to multi-dimensions. A mixture of this model is used as the wavelet coefficient prior for Bayesian wavelet based image denoising. The mixture model has a small number of parameters yet fits the marginal distribution of wavelet coefficients well. Despite being a stationary probability model, it is able to capture the dependencies among coefficients. Efficient parameter estimation methods and denoising rules are derived for the model. Denoising results are compared with existing techniques in both PSNR values and visual quality.