计算机应用研究2009，Vol.26Issue(8)：3134-3136,3.DOI:10.3969/j.issn.1001-3695.2009.08.097
基于小波域HMT模型的序列图像超分辨率重建
Super-resolution reconstruction of image sequence based on wavelet-HMT
摘要
Abstract
This paper proposed a super-resolution reconstruction algorithm of image sequence based on wavelet-domain hidden Markov tree (WHMT). Firstly, introduced the discrete wavelet transform briefly, wavelet-domain hidden Markov tree modeled the dependencies of multiscale wavelet coefficients through the state probabilities of the wavelet coefficients, whose distribution densities could be approximated by the Gaussian mixture. Secondly, from the Bayesian viewpoint and under the maximum a posteriori (MAP) probability estimation framework and exploiting the interrelated information between multiple warped, blurred, decimated and noisy low resolution (LR) images, concluded a minimum functional for the super-resolution reconstruction of image sequence using WHMT as the prior knowledge. Finally, adopted the expectation maximization (EM) algorithm and conjugate gradient (CG) algorithm to compute the HMT parameters and reconstruction image alternately. Experimental results indicate that the proposed algorithms is preferable in terms of both objective measurements and visual evaluation.关键词
超分辨率重建/序列图像/小波变换/隐马尔可夫树Key words
super-resolution reconstruction/image sequence/wavelet transform/HMT(hidden Markov tree)分类
信息技术与安全科学引用本文复制引用
周文婷,王庆..基于小波域HMT模型的序列图像超分辨率重建[J].计算机应用研究,2009,26(8):3134-3136,3.基金项目
国家教育部新世纪优秀人才支持计划基金资助项目(NCET-06-0882) （NCET-06-0882）
