Wavelet Domain Image Denoising Via Improved Hidden Markov Tree Model

摘要

The hidden Markov tree (HMT) model is a novel statistical model for image processing on wavelet domain. It can captures the persistence property of wavelet coefficients, but lack the clustering property of wavelet coefficients within a wavelet scale.The proposed contextual hidden Markov tree (CHMT) model enhances the clustering property of the HMT model by adding extended coefficients associated with wavelet coefficients without changing the wavelet tree structure. To avoid repeating computation, the upward-downward algorithm is used to train the CHMT model parameters. With the aid of empirical Bayesian estimation,the CHMT model is applied to remove noise from images corrupted with Gaussian noise. In experiments, the proposed CHMT model produced almost better results than the HMT model produced for image denoising. Furthermore, the CHMT model needs fewer iterations of training than the HMT model needs to get the same denoised results.