基于高斯比例混合模型的图像非下采样Contourlet域去噪

周汉飞; 王孝通; 徐晓刚

doi:10.3724/SP.J.1146.2008.00588

基于高斯比例混合模型的图像非下采样Contourlet域去噪

doi: 10.3724/SP.J.1146.2008.00588

周汉飞^{①③;王孝通},
王孝通,
徐晓刚

基金项目:

辽宁省自然科学基金(20062191)和浙江大学CADCG国家重点实验室开放基金资助课题

计量
- 文章访问数: 3550
- HTML全文浏览量: 112
- PDF下载量: 1027
- 被引次数: 0
出版历程
- 收稿日期: 2008-05-14
- 修回日期: 2009-03-26
- 刊出日期: 2009-08-19

Image Denoising Using Gaussian Scale Mixture Model in the Nonsubsampled Contourlet Domain

摘要

摘要: 为改善图像的去噪效果，该文提出了一种基于高斯比例混合模型的图像非下采样Contourlet域去噪算法。该算法首先建立非下采样Contourlet系数邻域的高斯比例混合模型，然后在模型基础上应用贝叶斯最小二乘法对系数进行估计，最后反变换得到恢复图像。算法结合了非下采样Contourlet变换对图像边缘的高效表示能力、非下采样变换的移不变性质以及GSM模型对非下采样Contourlet系数邻域相关性的概括能力。实验结果表明，该算法在视觉效果和峰值信噪比的改善上都取得了非常好的效果。
- 图像去噪;非下采样Contourlet变换;高斯比例混合
Abstract: A new method which using Gaussian scale mixtures model in the nonsubsampled Contourlet domain is proposed for image denoising. First, a Gaussian scale mixture model is introduced in order to capturing the correlation of nonsubsampled Contourlet locally coefficients. Then the coefficients are estimated by Bayes least squares estimator based on the model. Finally, the inverse nonsubsampled Contourlet transform is applied to the modified coefficients. This arithmetic combines the character of nonsubsampled Contourlet for image edge representation, shift-invariance and the effective of Gaussian scale mixture model for capturing correlation of locally coefficients. The numerical experimental results show the validity of the proposed method.

HTML全文

参考文献(1)

Dohono D L and Johnstone I M. Ideal spatial adaptation viawavelet shrinkage[J].Biometrika.1994, 81(3):425-455[2]Chang S G, Yu Bin, and Vetterli M. Adaptive waveletthresholding for image denoising and compression[J].IEEETransactions on Image Processing.2000, 9(9):1532-1546[3]Strela V, Portilla J, and Simonceil E P. Image denoising usinga local Gaussian scale mixture model in the waveletdomain[C]. Processings of SPIE Conference on WaveletApplications in Signal and Image Processing, San Diego, CA,2000, 4119: 363-371.[4]Portilla J, Strela V, and Wainwright M, et al.. Imagedenoisng using Gaussian Scale Mixtures in the WaveletDomain[J].IEEE Transactions on Image Processing.2003,12(11):1338-1351[5]田沛, 李庆周, 马平等. 一种基于小波变换的图像去噪新方法[J]. 中国图象图形学报, 2008, 13(3): 394-399.Tian Pei, Li Qing-zhou, and Ma Ping, et al.. A new methodbased on wavelet transform for image denoising[J]. Journal ofImages and Graphics, 2008, 13(3): 394-399.[6]Candes E J. Ridgelets: theory and application[D].[Ph.D.dissertation]. Department of Statistics, StandardUniversity, 1998.[7]Candes E J and Donoho D L. New tight frames of Curveletsand optimal representation of objects with smoothsingularities[R]. USA: Department of Statistics, StandardUniversity, 2002.[8]Starck J L, Candes E J, and Dohono D L. The Curvelettransform for image denoising[J].IEEE Transactions onImage Processing.2002, 11(6):670-684[9]Do M N. Directional multiresolution image representation[D].[Ph.D. dissertation]. School Comput. Commun. Sci. SwissFed. Inst. Technol. , Lausanne, Switzerland, 2001.[10]Do M N and Vetterli M. The Contourlet transform: Anefficient directional multiresolution image representation[J].IEEE Transactions on Image Processing.2005, 14(12):2091-2106[11]Starck J L, Donoho D L, and Candes E J. Very high qualityimage restoration by combining Wavelets and Curvelets[C].46th SPIE Annual Meeting, Conference on Signal and ImageProcessing, San Diego, USA, July 2001, 4478: 9-19.[12]金海燕, 焦李成, 刘芳. 基于Curvelet 域隐马尔科夫树模型的SAR 图像去噪[J]. 计算机学报, 2007, 30(3): 491-496.Jin Hai-yan, Jiao Li-cheng, and Liu Fang. SAR imagede-noising based on Curvelet domain hidden Markov treemodels[J]. Chinese Journal of Computers, 2007, 30(3):491-496.[13]Cunha A L, Zhou Jian-ping, and Do M N. Thenonsubsampled Contourlet transform: theory, design andapplication[J].IEEE Transactions on Image Processing.2006,15(10):3089-3101[14]Box G E P and Tiao C. Bayesian Inference in StatisticalAnalysis[M]. Reading, MA: Addison Wesley, 1992: 159-163.[15]Romberg J, Choi H, and Baraniuk P G. Bayesian waveletdomain image modeling using hidden Markov models[J].IEEE Transactions on Image Processing.2001, 10(7):1056-1068[16]Duncan D Y P and Do M N. Directional multiscale modelingof images using the Contourlet transform[J].IEEETransactions on Image Processing.2006, 15(6):1610-1620

施引文献

资源附件(0)

访问统计