基于图像块先验的低秩近似和维纳滤波的去噪算法

doi:10.6040/j.issn.1672-3961.0.2017.009

山东大学学报(工学版) ›› 2017, Vol. 47 ›› Issue (3): 16-20.doi: 10.6040/j.issn.1672-3961.0.2017.009

基于图像块先验的低秩近似和维纳滤波的去噪算法

张杨,陈飞^*,徐海平

福州大学数学与计算机科学学院, 福建福州 350116

收稿日期:2017-01-05 出版日期:2017-06-20 发布日期:2017-01-05
通讯作者: 陈飞(1980— ),男,福建福州人,副教授,博士,主要研究方向为图像处理与机器学习.E-mail:chenfei314@fzu.edu.cn E-mail:540615833@qq.com
作者简介:张杨(1992— ),女,福建三明人,硕士研究生,主要研究方向为图像/视频处理与分析.E-mail:540615833@qq.com
基金资助:
国家自然科学基金资助项目(61401098);福州大学科研启动基金资助项目(022575);福州大学科技发展基金资助项目(2014-XY-21)

Image patch prior based denoising algorithm by using low rank approximation and Wiener filtering

ZHANG Yang, CHEN Fei^*, XU Haiping

College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350116, Fujian, China

Received:2017-01-05 Online:2017-06-20 Published:2017-01-05

摘要/Abstract

摘要： 利用混合高斯模型(gaussian mixture model, GMM)学习自然图像块的纹理结构,提出一种基于图像块先验的低秩近似和维纳滤波的去噪算法。该算法能够同时利用外部图像块的先验结构信息和内部图像的自相似性,对待去噪图像进行分块聚类,并根据每类相似块的数量进行协同滤波。当相似图像块数量较多时,采用低秩近似的方法复原,有效利用图像的内部自相似性;当相似图像块数量较少时,采用维纳滤波,利用先验信息保持图像重要的纹理结构。试验结果表明此方法较适用于弧形边界和角点等存在较少相似块的自然图像,其峰值信噪比(peak signal to noise ratio, PSNR)和视觉效果优于目前部分主流算法。

关键词: 先验, 低秩近似, 维纳滤波, 高斯混合模型

Abstract: A Gaussian mixture model(GMM)was used to study the texture structure of natural image patches, and a low-rank approximation and Wiener filtering algorithm based on image patch prior were proposed. The proposed method divided the image into a number of overlapped patches and clustered them for collaborative filtering by using the prior structures of external image patch and internal image self-similarity. By grouping nonlocal similar patches, low-rank approximation was used as collaborative filtering to recover the texture structures. When the number of similar patches was small, Wiener filtering with patch prior was adopted to preserve texture features. The experimental results indicated that the proposed method was more suitable for the images with fewer similar patches like boundary and corner etc., and showed very competitive performance with state-of-the-art denoising method in terms of Peak Signal to Noise Ratio(PSNR)and visual quality.

Key words: low-rank approximation, prior, Wiener filtering, Gaussia mixture model

中图分类号:

TP37

张杨,陈飞,徐海平. 基于图像块先验的低秩近似和维纳滤波的去噪算法[J]. 山东大学学报(工学版), 2017, 47(3): 16-20.

ZHANG Yang, CHEN Fei, XU Haiping. Image patch prior based denoising algorithm by using low rank approximation and Wiener filtering[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(3): 16-20.

参考文献

[1] ZORAN D, WEISS Y. From learning models of natural image patches to whole image restoration[C] //Proceedings of the 13^th International Conference on Computer Vision(ICCV 2011). Barcelona, Spain: IEEE Computer Society, 2011, 6669(5): 479-486.
[2] CHATTERJEE P, MILANFAR P. Learning denoising bounds fornoisy images[C] //Proceedings of the 17^thInternational Conference on Image Processing(ICIP 2010). Hong Kong, China: IEEE Computer Society, 2010:1157-1160.
[3] RUDIN L I, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithms[J].Physica D: Nonlinear Phenomena,1992, 60(1):259-268.
[4] PORTILLA J, STRELA V, WAINWRIGHT M, et al, Image denoising using scale mixturesof Gaussians in the waveletdomain[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2003, 12(11):1338-1351.
[5] BUADES A, COLL B, MOREL J M. A nonlocal algorithm for image denoising[C] //Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR'05). SanDiego, USA: IEEE Computer Society, 2005.
[6] DABOV K, FOI A, KATKOVNIK V, et al. Image denoising by sparse 3-d transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007,16(8):2080-2095.
[7] MAIRAL J, BACH F, PONCE J, et al. Non-local sparse models for image restoration[C] //Proceedings of the 12th International Conference on Computer Vision(ICCV 2009). Kyoto, Japan: IEEE Computer Society, 2009, 30(2):2272-2279.
[8] AHARON M, ELAD M, BRUCKSTEIN A. K-svd: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322.
[9] DONG W, SHI G, LI X. Nonlocal image restoration wi-th bilateral variance estimation: a low-rank approach[J].IEEE Transaction on Image Processing, 2013, 22(2):700-711.
[10] GU S, ZHANG L, ZUO W, et al. Weighted nuclear nor-m minimization with application to image denoising[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR 2014). Columbus, USA: IEEE Computer Society, 2014:2862-2869.
[11] DONG W, SHI G, LI X, et al. Compressive sensing via nonlocal low rank regularization[J]. IEEE Transactions on Image Processing, 2014, 23(8): 3618-3632.
[12] 刘波,杨华,张志强.基于奇异值分解的图像去噪[J].微电子学与计算机,2007,24(11):169-171. LIU Bo, YANG Hua, ZHANG Zhiqiang. Image denoisingbased on singular value decomposition[J]. Microelectronics and Computer, 2007, 24(11):169-171.
[13] 张俊峰,孙清伟.基于图像旋转和分块的奇异值分解图像去噪[J].激光与红外,2009,39(5):538-541. ZHANG Junfeng, SUN Qingwei. Image denoising based on SVD using image rotation and block[J]. Laser and Infrared, 2009, 39(5):538-541.
[14] BURGER H, SCHULER C, HARMELING S. Image denoising: can plain neural networks compete with bm3d[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR 2012). Rhode Island, USA: IEEE Computer Society, 2012:2392-2399.
[15] CHEN F, ZHANG L, YU H M. External patch prior guided internal clustering for image denoising[C] //Proceedings of the 2015 IEEE International Conference on Computer Vision(ICCV). San Diego, USA: IEEE Computer Society, 2015:603-611.
[16] DEMPSTER A P, LAIRD N M, RUBIN D B.Maximum likelihood from incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society. Series B(Methodological), 1977, 39(1):1-38.
[17] CAI J F, CANDES E J, SHEN Z W. A singular value thresholding algorithm for matrix completion[J]. SIAM Journal on Optimization, 2010, 20(4):1956-1982.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于图像块先验的低秩近似和维纳滤波的去噪算法

Image patch prior based denoising algorithm by using low rank approximation and Wiener filtering

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

多维度评价

本文评价

推荐阅读 10

[1]	徐芊芊,许倩,徐华畅,赵钰琳,徐凯,朱红. 基于CnViT的胶质瘤IDH1突变状态智能预测方法[J]. 山东大学学报 (工学版), 2023, 53(2): 127-134.
[2]	杨霄,袭肖明,李维翠,杨璐. 基于层次化双重注意力网络的乳腺多模态图像分类[J]. 山东大学学报 (工学版), 2022, 52(3): 34-41.
[3]	赵加敏,冯爱民*,刘学军. 局部密度嵌入的结构单类支持向量机[J]. 山东大学学报(工学版), 2012, 42(4): 13-18.
[4]	张友新,王立宏. 两阶段近邻传播半监督聚类算法[J]. 山东大学学报(工学版), 2012, 42(2): 18-22.
[5]	梁浩哲,徐树奎,李国辉,张军. 面向监控视频的行为模式挖掘[J]. 山东大学学报(工学版), 2011, 41(6): 24-30.
[6]	陈涛,方志刚,徐洁 . 基于人脸和语音的混合型身份认证系统[J]. 山东大学学报(工学版), 2008, 38(2): 56-60 .
[7]	崔英,陈文楷,雷飞 . 基于背景减法的游泳者检测[J]. 山东大学学报(工学版), 2008, 38(1): 39-42 .