基于字典学习的非局部均值去噪算法

doi:10.3724/SP.J.1087.2013.01420

计算机应用 ›› 2013, Vol. 33 ›› Issue (05): 1420-1422.DOI: 10.3724/SP.J.1087.2013.01420

基于字典学习的非局部均值去噪算法

崔学英¹,²,张权²,桂志国²

1. 太原科技大学应用科学学院，太原 030024
2. 中北大学电子测试技术国防重点实验室，太原 030051

收稿日期:2012-11-08 修回日期:2012-12-24 出版日期:2013-05-01 发布日期:2013-05-08
通讯作者: 崔学英
作者简介:崔学英(1978-)，女，山西临汾人，讲师，博士研究生，主要研究方向：图像处理、微分方程应用；张权(1974-)，男，山西大同人，讲师，博士研究生，主要研究方向：图像处理、科学可视化；桂志国(1972-)，男，河北蓟县人，教授，博士，主要研究方向：无损检测、图像处理。
基金资助:
国家自然科学基金资助项目(61202311);山西省自然科学基金资助项目(2009011020-2);山西省研究生优秀创新项目(20123098);太原科技大学校青年基金资助项目(20113019)

Non-local means denoising approach based on dictionary learning

CUI Xueying¹,²,ZHANG Quan¹,GUI Zhiguo¹

1. National Key Laboratory for Electronic Measurement Technology, North University of China, Taiyuan Shanxi 030051,China
2. School of Applied Science, Tauyuan University of Science and Technology, Taiyuan Shanxi 030024, China

Received:2012-11-08 Revised:2012-12-24 Online:2013-05-08 Published:2013-05-01
Contact: CUI Xueying

摘要/Abstract

摘要： 针对非局部均值中相似度的衡量问题,提出了一种基于字典学习的度量算法。首先利用局部像素群块匹配方法消除不相似的图像块带来的干扰,然后对含有噪声的相似块采用字典学习的方法降噪。与经典的字典学习不同的是,对相似块采用联合稀疏编码的思想,利用主成分分析法学习一个高效紧字典,保留相似块间的相关性信息。采用降噪后的图像块间的欧氏距离计算像素间的相似度,能更好地反映相似块的相似性。实验结果表明,所提出的方法在峰值信噪比和视觉效果方面都优于传统算法,尤其对含有较多细节且结构相似性强的图像,细节和纹理部分的保持效果更好,算法的鲁棒性也优于传统算法.

关键词: 图像去噪, 字典学习, 主成分分析, 稀疏表示, 非局部均值

Abstract: Concerning the measurement of the similarity of non-local means, a method based on dictionary learning was presented. First, block matching based local pixel grouping was used to eliminate the interference by dissimilar image blocks. Then, the corrupted similar blocks were denoised by dictionary learning. As a further development of classical sparse representation model, the similar patches were unified for joint sparse representation and learning an efficient and compact dictionary by principal component analysis, so that the similar patches relevency could be well preserved. This similarity between the pixels was measured by the Euclidean distance of denoised image blocks,which can well show the similarity of the similar blocks. The experimental results show the modified algorithm has a superior denoising performance than the original one in terms of both Peak Signal-to-Noise Ratio (PSNR) and subjective visual quality. For some images whose structural similarity is large and with rich detail information, their structures and details are well preserved. The robustness of the presented method is superior to the original one.

Key words: image denoising, dictionary learning, Principal Component Analysis (PCA), sparse representation, non-local means

中图分类号:

TP317.4

崔学英张权桂志国. 基于字典学习的非局部均值去噪算法[J]. 计算机应用, 2013, 33(05): 1420-1422.

CUI Xueying ZHANG Quan GUI Zhiguo. Non-local means denoising approach based on dictionary learning[J]. Journal of Computer Applications, 2013, 33(05): 1420-1422.

参考文献

［1］ELAD M, AHARON M. Image denoising via sparse and redundant representations over learned dictionaries［J］. IEEE Transactions on Image Processing, 2006,15(12):3736-3745.

［2］AHARON M, ELAD M, BRUCKSTEIN A M. The K-SVD: an algorithm for designing of overcomplete dictionaries for sparse representation［J］. IEEE Transactions on Signal Processing,2006,54(11):4311-4322.

［3］BUADES A, COLL B, MOREL J M. A review of image denoising algorithms, with a new one［J］. Multiscale Modeling and Simulation, 2005, 4(2):490-530.

［4］BUADES A, COLL B, MOREL J M.Nonlocal image and movie denoising［J］. International Journal of Computer Vision，2008，76(2)：123-139．

［5］李民,程建,李小文,等.非局部学习字典的图像修复［J］.电子与信息学报,2011,88(11): 2672-2678.

［6］DONG W S, ZHANG L, SHI G M,et al. Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization［J］. IEEE Transactions on Image Processing, 2011,20(7):1838-1857.

［7］ELAD M, YAVNEH I. A plurality of sparse representation is better than the sparsest one alone［J］. IEEE Transactions on Information Theory, 2009,55(10):4701-4714.

［8］PROTTER M, YAVNEH I, ELAD M. Closed-form MMSE estimation for signal denoising under sparse representation modeling over a unitary dictionary［J］. IEEE Transactions on Signal Processing, 2010,58(7):3471-3484.

［9］MAIRAL J, BACH F, PONCE J,et al.Non-local sparse models for image restoration［C］// 2009 IEEE 12th International Conference on Computer Vision. New York:IEEE, 2009:2272-2279.

［10］TROPP J A. Algorithms for simultaneous sparse approximation［J］. Signal Processing, 2006,86(3):572-602.

［11］刘晓明,田雨,何徽,等.一种改进的非局部均值图像去噪算法［J］.计算机工程, 2012, 38(4):199-201.

［12］ZHANG L, DONG W S, ZHANG D, et al.Two-stage image denoising by principal component analysis with local pixel grouping［J］. Pattern Recognition, 2010,43(4): 1531-1549.

［13］TASDIZEN T. Principal components for non-local means image denoising［C］// 15th International Conference on Image Processing. New York: IEEE, 2008：1728-1731.

[1]	裴春阳, 樊宽刚, 马政. 基于边缘保留分解和改进稀疏表示的医学图像融合[J]. 计算机应用, 2021, 41(7): 2092-2099.
[2]	王心, 朱浩华, 刘光灿. 卷积鲁棒主成分分析[J]. 计算机应用, 2021, 41(5): 1314-1318.
[3]	任晓奎, 刘鹏飞, 陶志勇, 刘影, 白立春. 基于单快拍信号到达角估计算法的室内入侵检测[J]. 计算机应用, 2021, 41(4): 1153-1159.
[4]	陆荣秀, 陈明明, 杨辉, 朱建勇. 基于溶液图像时序特征的元素组分含量动态监测系统[J]. 计算机应用, 2021, 41(10): 3075-3081.
[5]	王丽娟, 陈少敏, 尹明, 许跃颖, 郝志峰, 蔡瑞初, 温雯. 基于近邻图改进的块对角子空间聚类算法[J]. 计算机应用, 2021, 41(1): 36-42.
[6]	陈利霞, 班颖, 王学文. 基于张量核范数与3D全变分的背景减除[J]. 计算机应用, 2020, 40(9): 2737-2742.
[7]	郑延斌, 韩梦云, 樊文鑫. 基于二维主成分分析与卷积神经网络的手写体汉字识别[J]. 计算机应用, 2020, 40(8): 2465-2471.
[8]	李东博, 黄铝文. 重加权稀疏主成分分析算法及其在人脸识别中的应用[J]. 计算机应用, 2020, 40(3): 717-722.
[9]	高彦彦, 李莉, 张晶, 贾英茜. 结合梯度投影稀疏重构和复数小波的图像重构[J]. 计算机应用, 2020, 40(2): 486-490.
[10]	王海鹏, 降爱莲, 李鹏翔. 牛顿-软阈值迭代鲁棒主成分分析算法[J]. 计算机应用, 2020, 40(11): 3133-3138.
[11]	宗春梅, 张月琴, 曹建芳, 赵青杉. 基于深度先验及非局部相似性的压缩感知核磁共振成像[J]. 计算机应用, 2020, 40(10): 3054-3059.
[12]	张晓博, 杨燕, 李天瑞, 陆凡, 彭莉兰. 基于医疗文本数据聚类的帕金森病早期诊断预测[J]. 计算机应用, 2020, 40(10): 3088-3094.
[13]	陈莉, 陈晓云. 结合局部熵和鲁棒主成分分析的眼底图像硬性渗出物检测方法[J]. 计算机应用, 2019, 39(7): 2134-2140.
[14]	周非, 夏鹏程. 基于主成分分析和卡方距离的信号强度差指纹定位算法[J]. 计算机应用, 2019, 39(5): 1405-1410.
[15]	徐涛, 王晓明. 泛化误差界指导的鉴别字典学习[J]. 计算机应用, 2019, 39(4): 940-948.

基于字典学习的非局部均值去噪算法

Non-local means denoising approach based on dictionary learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics