CASIA OpenIR  > 智能制造技术与系统研究中心  > 多维数据分析
汪雪林; 赵书斌; 彭思龙; WANGXue-lin; ZHAOShu-Bin; PENGSi-long,
Source Publication计算机学报
Volume28(6) (EI)Issue:2005年06期Pages:1006-1012
Other AbstractFrom the viewpoint of Bayesian method for image restoration,a iinear image restoration aigorithm based on waveiet domain Hidden Markov Tree(HMT)modei is proposed. Waveiet-domain HMT modeis the dependencies of muitiscaie waveiet coefficients through the state probabiiities of the waveiet coefficients,whose distribution densities can be approximated by Gaussian mixture modei. The proposed aigorithm specifies the prior distribution of reai-worid images through waveiet-domain HMT modei and converts the restoration probiem to an constrained optimization task which can be soived with the steepest descend method. Parameters of the HMT modei are adaptiveiy determined through a fast estimation method which avoids the time-consuming training process. Experimentai resuits show that the aigorithm properiy retrieves various kinds of edges and the PNSR and subjective visuai effect of the restored images are improved significantiy.
Keyword图像复原 / 小波变换 / 隐马尔可夫树模型 / 最速下降法
Document Type期刊论文
Corresponding Author汪雪林
Recommended Citation
GB/T 7714
汪雪林,赵书斌,彭思龙,等. 基于小波域隐马尔可夫树模型的图像复原[J]. 计算机学报,2005,28(6) (EI)(2005年06期):1006-1012.
APA 汪雪林,赵书斌,彭思龙,WANGXue-lin,ZHAOShu-Bin,&PENGSi-long,.(2005).基于小波域隐马尔可夫树模型的图像复原.计算机学报,28(6) (EI)(2005年06期),1006-1012.
MLA 汪雪林,et al."基于小波域隐马尔可夫树模型的图像复原".计算机学报 28(6) (EI).2005年06期(2005):1006-1012.
Files in This Item:
File Name/Size DocType Version Access License
基于小波域隐马尔可夫树模型的图像复原.p(503KB)期刊论文作者接受稿暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[汪雪林]'s Articles
[赵书斌]'s Articles
[彭思龙]'s Articles
Baidu academic
Similar articles in Baidu academic
[汪雪林]'s Articles
[赵书斌]'s Articles
[彭思龙]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[汪雪林]'s Articles
[赵书斌]'s Articles
[彭思龙]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.