Wavelet Domain Deblurring and Denoising for Image Resolution Improvement

In this paper, a new image interpolation method which is combined with deblurring and denoising is proposed. The MAP (Maximum a Posteriori) estimate is adopted to deal with the ill-conditioned problem (obtaining a super resolution image from a sub-sampled, blurred and contaminated image) in the wavelet domain. The universal hidden Markov tree (uHMT) theory in the wavelet domain is applied to construct a prior model for the MAP estimate. The results show that images reconstructed by our method are much better and sharper than those recovered images by the Huber- Markov random field (HMRF) prior model for MAP in the space domain.