Image Denoising Using Bandelets and HiddenMarkov Tree Models

In this paper, both the marginal and jointstatistics of second generation Orthogonal bandelet transform(OBT) coefficients of natural images are firstly studied,and the highly non-Gaussian marginal statistics andstrong interscale, interlocation and interdirection dependenciesamong OBT coefficients are found. Then a HiddenMarkov tree (HMT) model in OBT domain which can effectivelycapture all dependencies across scales, locationsand directions is developed. The main contribution of thispaper is that it exploits the edge direction information ofOBT coefficients, and proposes an image denoising algorithm(B-HMT) based on HMT model in OBT domain.We apply B-HMT to denoise natural images which contaminatedby additive Gaussian white noise, and experimentalresults show that B-HMT outperforms the Wavelet HMT(W-HMT) and Contourlet HMT (C-HMT) in terms of visualeffect and objective evaluation criteria.