Bayesian denoising of visual images in the wavelet domain
Jie Jia · Journal of Xi'an University of Post and Telecommunications · 2003
Denoising is one of the key problems of image processing.Recent research suggests that wavelet coefficients for natural images tend to have a Laplacian distribution function.If wavelet transform used is orthogonal,noises in wavelet coefficients of those images corrupted by additive white Gaussian noises are also Gaussian and white.Image denoising can be realized by filtering the coefficients based on Bayesian estimator.This paper compares the linear estimator and Bayesian estimator in noise reduction.Simulation results show that the latter has better performance.