Image denoising employing a bivariate pearson distribution with rayleigh density priori for statistical parameter

In wavelet-based Bayesian denoising, the performance of several methods strongly depends on the correctness of the distribution that is used to describe the data. Therefore, the selection of a proper model for distribution is thus an important issue in the denoising process. This paper presents a new image denoising algorithm based on bivariate Pearson type VII distribution with approximated MAP estimation for statistical parameter (local variance) using Rayleigh density priori for observed variance and Gaussian distribution for noisy wavelet coefficients. The bivariate probability density function (pdf) takes into account the statistical dependency among wavelet coefficients, the local variation and the correlation between the coefficient amplitudes. The experimental results show that the proposed technique outperforms several exiting methods both visually and in terms of peak signal-to-noise ratio (PSNR).