非局部MCMC采样和低秩逼近的图像去噪算法

doi:10.3969/j.issn.1001-2400.2013.06.024

J4 ›› 2013, Vol. 40 ›› Issue (6): 140-146.doi: 10.3969/j.issn.1001-2400.2013.06.024

非局部MCMC采样和低秩逼近的图像去噪算法

罗亮¹;冯象初¹;霍雷刚¹;张选德^1,2;吴玉莲^1,3;李小平¹

(1. 西安电子科技大学理学院，陕西西安 710071；
2. 宁夏大学数学计算机学院，宁夏银川 750021;
3. 西安医学院公共课部，陕西西安 710021)

收稿日期:2012-08-28 出版日期:2013-12-20 发布日期:2014-01-10
通讯作者: 罗亮
作者简介:罗亮(1982-)，男，西安电子科技大学博士研究生，E-mail: luoliang775@163.com．
基金资助:
国家自然科学基金资助项目(61271294,60872138, 61105011, 11101292)

Image denoising method based on non-local Markov-chain Monte Carlo sampling and low rank approximation

LUO Liang¹;FENG Xiangchu¹;HUO Leigang¹;ZHANG Xuande^1,2;WU Yulian^1,3;LI Xiaoping¹

(1. School of Science, Xidian Univ., Xi'an 710071， China;
2. School of Mathematics and Computer, Ningxia Univ., Yingchuan 750021， China;
3. Common Course Department, Xi'an Medical College, Xi’an 710021， China)

Received:2012-08-28 Online:2013-12-20 Published:2014-01-10
Contact: LUO Liang

摘要/Abstract

摘要：

针对噪声把原图像中的一些细节掩盖了的问题，提出一种非局部马尔科夫蒙特卡罗采样和低秩逼近的随机去噪方法．首先通过马尔科夫蒙特卡罗随机采样寻找每个图像块的相似匹配块簇，然后对这些相似匹配块簇进行奇异值分解，用分解后的低秩结构恢复原图像，从而达到去噪的目的．实验表明，这种方法计算复杂度低．与非局部平均方法相比，较好地保留了边缘等细节信息; 与BM3D方法相比，能保持较好的视觉质量．

关键词: 图像去噪, 非局部马尔科夫蒙特卡罗方法, 低秩逼近, 后验概率估计

Abstract:

Combining non-local Markov-chain Monte Carlo sampling and low-rank approximation of the matrix method, an approach for image noise removal is presented. The cluster of similar patches is searched by using Markov-chain Monte Carlo sampling. The cluster matrix of similar patches is decomposed by the singular value decomposition method,and the image noise is suppressed by applying the low rank structure from decomposing. Simulation results show that the proposed method outperforms the BM3D and the non-local means (NLM)method in computational-complexity. The proposed method has a better performance in protecting image details compared with the NLM method, and has some advantages over the BM3D method in terms of visual quality.

Key words: image denoising, non-local Markov-Chain Monte Carlo method, approximation of low rank matrix, posterior probability estimate

中图分类号:

TP391

罗亮;冯象初;霍雷刚;张选德;吴玉莲;李小平. 非局部MCMC采样和低秩逼近的图像去噪算法[J]. J4, 2013, 40(6): 140-146.

LUO Liang;FENG Xiangchu;HUO Leigang;ZHANG Xuande;WU Yulian;LI Xiaoping. Image denoising method based on non-local Markov-chain Monte Carlo sampling and low rank approximation[J]. J4, 2013, 40(6): 140-146.

参考文献

［1］王卫卫, 席灯炎, 杨塨鹏, 等. 利用结构纹理分解的海洋舰船目标检测［J］. 西安电子科技大学学报, 2012, 39(4): 131-137.
Wang Weiwei, Xi Dengyan, Yang Gongpeng, et al. Warship Detection Algorithm Based on Cartoon-texture Decomposition［J］. Journal of Xidian University, 2012, 39(4): 131-137.
［2］韩雨, 王卫卫, 冯象初. 基于迭代重加权的非刚性图像配准［J］. 自动化学报, 2011, 37(9): 1059-1066.
Han Yu, Wang Weiwei, Feng Xiangchu. Iteratively Reweighted Method Based Nonrigid Image Registration［J］. Acta Automatica Sinica, 2011, 37(9): 1059-1066.
［3］侯彪, 徐婧, 刘凤, 等. 基于第二代Bandelet域隐马尔科夫树模型的图像分割［J］. 自动化学报, 2009, 35(5): 498-504.
Hou Biao, Xu Jing, Liu Feng, et al. Image Segmentation Using Second Generation Bandelet-domain Hidden Markov Treemodels［J］. Acta Automatica Sinica, 2009, 35(5): 498-504.
［4］ Katkovnik V, Foi A, Egiazarian K, et al. From Local Kernel to Nonlocal Multiple-Model Image Denoising［J］. International Journal of Computer Vision, 2010, 86(1): 1-32.
［5］ Lindenbaum M, Fischer M, Bruckstein A. On Gabor's Contribution to Imageenhancement［J］. Pattern Recognition, 1994, 27: 1-8.
［6］ Bemstein R. Adaptive Nonlinear Filters for Simultaneous Removal of Different Kinds of Noise in Images［J］. IEEE Transactions on Circuits and Systems, 1987, 34(11): 1275-1291.
［7］ Wiener N. Extrapolation, Interpolation, and Smoothing of Stationary Time Series［M］. New York: Wiley, 1949.
［8］ Efros A, Leung T. Texture Synthesis By Non Parametric Sampling［C］//Proceedings of the IEEE International Conference on Computer Vision. Piscataway: IEEE Computer Society, 1999: 1033-1038.
［9］ Wong A, Mishra A K, Zhang Wen, et al. Stochastic Image Denoising Based on Markov-chain Monte Carlo Sampling［J］. Signal Processing, 2011, 91(8): 2112-2120.
［10］ Fieguth P. Statistical Image Processing and Multidimensional Modeling［M］. Berlin: Springer, 2010.
［11］ Lee J S. Speckle Suppression and Analysis for Synthetic Aperture Radar Images［J］. Optical Engineering, 1986, 25(5): 636-643.
［12］ Hastings W. Monte Carlo Sampling Methods Using Markov Chains and Their Applications［J］. Biometrika, 1970, 57(1): 97-109.
［13］ Buades A, Coll B, Morel J. Nonlocal Image and Movie Denoising［J］. International Journal of Computer Vision, 2008, 76(2): 123-139.
［14］ Dabov K, Foi A, Katkovnik V, et al. Image Denoising By Sparse 3D Transform-domain Collaborative Filtering［J］. IEEE Transactions on Image Processing, 2007, 16(8): 2080-2095.

[1]	乔鱼,冯象初. 博弈方法下的图像去噪与边界提取[J]. 西安电子科技大学学报, 2021, 48(4): 128-135.
[2]	赖睿,官俊涛,徐昆然,熊皑,杨银堂. 级联残差学习的红外图像非均匀性校正方法[J]. 西安电子科技大学学报, 2019, 46(1): 14-19.
[3]	张瑞;冯象初;杨丽霞;常莉红. 全局稀疏梯度耦合张量扩散的图像去噪模型[J]. 西安电子科技大学学报, 2017, 44(6): 150-155.
[4]	冯象初;李晓晖;王卫卫;贾西西. 方向扩散方程修正BM3D图像去噪改进算法[J]. 西安电子科技大学学报, 2017, 44(5): 102-108.
[5]	霍雷刚;冯象初;王旭东;霍春雷. 高阶SVD和全变差正则的乘性噪声去除模型[J]. 西安电子科技大学学报, 2016, 43(3): 78-84.
[6]	高博;王俊;原慧. 一种参数自适应的SAR图像去噪方法[J]. J4, 2015, 42(5): 63-67.
[7]	马洪晋;聂玉峰. 一种图像特征增强的各向异性扩散去噪方法[J]. J4, 2015, 42(5): 154-160.
[8]	崔华;严玍伻. 融合卡方累积分布函数和PDE的曲波阈值去噪法[J]. J4, 2014, 41(4): 108-110+172.
[9]	王旭东;冯象初;张选德. 去除乘性噪声的迭代重加权二阶正则模型[J]. J4, 2014, 41(2): 130-136.
[10]	赵扬扬;周水生;武亚静. 一种用于人脸识别的非迭代GLRAM算法[J]. J4, 2014, 41(2): 144-150.
[11]	白键;冯象初. 去除乘性噪声的积分微分方程模型[J]. J4, 2013, 40(3): 132-138.
[12]	王旭东;冯象初;白键. 基于结构张量场拟合的图像恢复方法[J]. J4, 2011, 38(6): 68-74.
[13]	张伟斌；冯象初；王卫卫. 带Besov忠诚项的图像去噪变分模型[J]. J4, 2011, 38(3): 155-158+163.
[14]	张元科；张军英；卢虹冰. 基于参数估计的低剂量CT图像统计去噪[J]. J4, 2011, 38(3): 99-106.
[15]	冯象初;刘涛;李亚峰. 小波域中的广义非局部平均去噪算法[J]. J4, 2010, 37(5): 941-946.

非局部MCMC采样和低秩逼近的图像去噪算法

Image denoising method based on non-local Markov-chain Monte Carlo sampling and low rank approximation

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0