保持纹理细节的自适应非局部均值图像降噪

doi:10.3969/j.issn.1007-2861.2013.07.023

上海大学学报(自然科学版) ›› 2014, Vol. 20 ›› Issue (1): 99-106.doi: 10.3969/j.issn.1007-2861.2013.07.023

保持纹理细节的自适应非局部均值图像降噪

陈刚, 钱振兴, 王朔中

上海大学通信与信息工程学院, 上海 200444

收稿日期:2013-02-06 出版日期:2014-02-28 发布日期:2014-02-28
通讯作者: 王朔中(1943—), 男, 教授, 博士生导师, 博士, 研究方向为数字图像处理. E-mail:shuowang@shu.edu.cn
基金资助:
国家自然科学基金资助项目(61071187, 61103181); 上海市教委创新基金资助项目(11YZ10)

Adaptive Nonlocal Means Image Denoising for Better Preservation of Textures

CHEN Gang, QIAN Zhen-xing, WANG Shuo-zhong

School of Communication and Information Engineering, Shanghai University, Shanghai 200444, China

Received:2013-02-06 Online:2014-02-28 Published:2014-02-28

摘要/Abstract

摘要： 主邻域字典(principal neighborhood dictionaries, PND)非局部均值(nonlocal means, NLM)是一种基于主成分分析(principal component analysis , PCA)的有效图像降噪方法, 但因其未能充分利用图像的内容结构信息, 对纹理细节较多区域的降噪效果较差. 改进PND 方法, 实现基于PCA 的自适应非局部均值降噪. 根据图像局部内容调整滤波参数h, 得到动态变化的像素间相似权值. 实验结果表明, 该方法能更好地保留图像纹理和边缘信息, 降噪效果优于非自适应的PND 方法.

关键词: 非局部均值降噪, 滤波参数, 主成分分析, 主邻域字典

Abstract: The principal neighborhood dictionaries (PND) nonlocal means (NLM) is an effective method for image denoising, which is based on principal component analysis(PCA). However, it does not make full use of the image contents and is less effective in texture regions. This paper proposes modification to the PND method by adjusting a filter parameter h to achieve better accuracy. Experimental results show that the proposed outperforms PND method and can preserve more edge and texture details while achieving satisfactory denoising results.

Key words: filter parameter, nonlocal means (NLM) denoising, principal component analysis(PCA), principal neighborhood dictionaries (PND)

中图分类号:

TP 391

陈刚, 钱振兴, 王朔中. 保持纹理细节的自适应非局部均值图像降噪[J]. 上海大学学报(自然科学版), 2014, 20(1): 99-106.

CHEN Gang, QIAN Zhen-xing, WANG Shuo-zhong. Adaptive Nonlocal Means Image Denoising for Better Preservation of Textures[J]. Journal of Shanghai University（Natural Science Edition）, 2014, 20(1): 99-106.

参考文献

[1] 王锐玲, 施俊. 结合Tetrolet与主动随机场模型的高斯噪声抑制[J]. 上海大学学报: 自然科学版, 2012, 30(3): 275-280.

[2] Buades A, Coll B, Morel J. A review of image denoising algorithms [J]. Multiscale Modeling and Simulation, 2005, 4(2): 490-530.

[3] Buades A, Coll B, Morel J. A non-local algorithm for image denoising [C]//IEEE Conference on Computer Vision and Pattern Recognition. 2005: 60-65.

[4] Kervrann C, Boulanger J, Coupe P. Bayesian nonlocal means filter, image redundancy and adaptive dictionaries for noise removal [J]. Scale Space Methods Variational Methods in

Computer Vision, 2007, 4485: 520-532.

[5] Deledalle C, Denis L, Tupin F. Iterative weighted maximum likelihood denoising with probabilistic patch-based weights [J]. IEEE Transactions on Image Processing, 2009, 18(12): 2661-2672.

[6] Tasdizen T. Principal components for non-local means image denoising [C]//IEEE International Conference on Image Processing. 2008: 1728-1731.

[7] Tasdizen T. Principal neighborhood dictionaries for nonlocal means image denoising [J]. IEEE Transactions on Image Processing, 2009, 18(12): 2649-2660.

[8] van de Ville D, Kocher M. SURE-based nonlocal means [J]. IEEE Signal Process Letters, 2009, 16(11): 973-976.

[9] Salmon J. On two parameters for denoising with nonlocal means [J]. IEEE Signal Process Letters, 2010, 17(3): 269-272.

[10] Yan R, Shao L. Improved nonlocal means based on pre-classification and invariant block matching [J]. IEEE Transactions on Display Technology, 2012, 8(4): 212-218.

[11] 赖新平, 胡海平. 基于各向异性扩散的降噪新方法[J]. 上海大学学报: 自然科学版, 2009, 15(4): 346-350.

[12] Esbensen K, Geladi P. Principal component analysis [C]//Proceedings of the Multivariate Statistical Workshop for Geologists and Geochemists. 1987: 37-52.

[13] Wang S, Lu X, Su S. Image block feature vectors based on a singular-value information metric and color-texture description [J]. Journal of Shanghai University: English Edition, 2007, 11(3): 205-209.

保持纹理细节的自适应非局部均值图像降噪

Adaptive Nonlocal Means Image Denoising for Better Preservation of Textures

PDF

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 3

编辑推荐

Metrics

本文评价

[1]	刘冬军, 徐美华, 龚露鸣, 夏臣君. 前方车辆检测的特征融合算法研究与实现[J]. 上海大学学报(自然科学版), 2017, 23(6): 893-.
[2]	于冰1,2, 丁友东1,2, 董荪1,2, 黄曦1,2. 基于分组鲁棒主成分分析的老电影修复[J]. 上海大学学报(自然科学版), 2017, 23(3): 315-323.
[3]	钱权,陈孟,桂林开. 特征约简在入侵检测上的应用[J]. 上海大学学报(自然科学版), 2010, 16(1): 86-90.