首页 > 论文 > 光子学报 > 43卷 > 3期(pp:310001--1)

利用核模糊聚类和正则化的图像稀疏去噪

Image Denoising Using Kernel Fuzzy Clustering and Regularization on Sparse Model

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

针对目前图像去噪方法噪音抑制不彻底、容易模糊细节等问题,提出了一种利用核模糊C均值聚类和正则化的图像稀疏去噪方法.该方法首先将图像分成大小相同的若干块,并采用核模糊C均值聚类算法对相似的图像块进行聚类,从而保证同一类图像块共享相同的稀疏去噪模型;然后,选择由经典图像库中图像训练而得的全局字典作为初始字典,很好地适应图像的多种特征;接着,对于同一类图像块,通过施加1/2范数正则化约束,实现该类图像块在字典下的稀疏分解,确保分解系数更为稀疏;最后,通过改进的K-奇异值分解算法完成字典的更新,并选择与原稀疏模型差异最大的图像块来替换更新字典的冗余原子,从而有效地去除图像噪音.实验结果表明,与小波扩散去噪法、固定字典去噪法、最优方向去噪法、K-奇异值分解去噪法相比,该方法能更有效地去除图像噪音,保留图像细节,改善图像视觉效果.

Abstract

Aimed at the problems that the existing denoising methods suppress noise incompletely and blur the details of image, an image denoising method using kernel fuzzy C-means clustering and regularization on sparse model was proposed. Firstly, the image was divided into equal pieces and kernel fuzzy C-means clustering algorithm was used for clustering the similar image pieces, thereby ensuring image pieces in the same class share the same sparse denoising model. Then, the global dictionary trained by images from the classical image library was selected as the initial dictionary to adapt to the various characteristics of image very well. Next, a 1/2 norm regularization constraint condition was imposed and sparse decomposition of image pieces in the same class under the dictionary was achieved, which made decomposition coefficients sparser. Finally, the update of dictionary was completed by improved K-singular value decomposition algorithm, and image pieces with the largest difference from the original sparse model were selected to replace the redundancy atoms of the uapdated dictionary. Thus, noise in the image was suppressed effectively. Experimental results show that, compared with denoising method based on wavelet combining with nonlinear diffusion, denoising method based on constant dictionary, denoising method of optimal directions and K-singular value decomposition denoising method, the proposed method can remove noise of the image more effectively and preserve the details of the image and improve the visual effect better.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TN911.73

DOI:10.3788/gzxb20144303.0310001

基金项目:国家自然科学基金 (No. 60872065)和江苏高校优势学科建设工程资助

收稿日期:2013-07-22

修改稿日期:2013-11-18

网络出版日期:--

作者单位    点击查看

吴一全:南京航空航天大学 电子信息工程学院, 南京 210016
李立:南京航空航天大学 电子信息工程学院, 南京 210016

联系人作者:吴一全(nuaaimage@163.com)

备注:吴一全(1963-),男,教授,博士,博导,主要研究方向为红外目标检测与识别、遥感图像处理、视觉检测与图像测量、视频编码与视频处理、数字全息技术等.

【1】XU Shu-hua, QI Ming-ming. Image denoising based on multi-scales total least squares[J]. Acta Photonica Sinica, 2010, 39(5): 956-960.
许淑华, 齐鸣鸣. 基于多尺度总体最小二乘的图像去噪[J]. 光子学报, 2010, 39(5): 856-960.

【2】WU Yi-quan, JI Shou-xin. Image contourlet threshold de-noising based on chaotic particle swarm optimization[J]. Acta Photonica Sinica, 2010, 39(9): 1645-1651.
吴一全, 纪守新. 基于混沌粒子群优化的图像Contourlet阈值去噪[J]. 光子学报, 2010, 39(9): 1645-1651.

【3】WU Yi-quan, HOU Wen, WU Shi-hua. Image de-noising based on complex contourlet transformand nonlinear diffusion[J]. Journal of Circuits and Systems, 2012, 17(6): 111-116.
吴一全, 侯雯, 吴诗婳. 基于复Contourlet域非线性扩散的图像去噪[J]. 电路与系统学报, 2012, 17(6): 111-116.

【4】XIE Kai, ZHANG Feng. Image denoising method based on dictionary learning with union of two orthonormal bases[J]. Journal of Computer Applications, 2012, 32(4): 1119-1121.
解凯, 张芬. 基于双正交基字典学习的图像去噪方法[J]. 计算机应用, 2012, 32(4): 1119-1121.

【5】DENG Cheng-zhi. Image denoising algorithm based on multiscale ridgelet dictionary[J]. Computer Engineering, 2010, 36(23): 207-209.
邓承志. 基于多尺度脊波字典的图像去噪算法[J]. 计算机工程, 2010, 36(23): 207-209.

【6】ZENG Jun-ying, GAN Jun-ying, ZHAI Yi-kui. Face recognition based on fast sparse representation of Gabor dictionary and l0 norm[J]. Journal of Signal Processing, 2013, 29(2): 256-261.
曾军英, 甘俊英, 翟懿奎. Gabor字典及l0范数快速稀疏表示的人脸识别算法[J]. 信号处理, 2013, 29(2): 256-261.

【7】AHARON M, ELAD M, BRUCKSTEIN A. K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transations on Image Processing, 2006, 54(11): 4311-4322.

【8】TROPP J A, GIBERT A C. Signal recovery from random measurements via orthogonal matching pursuit[J]. IEEE Transactions on Information Theory, 2007, 53(12): 4655-4666.

【9】JOST P V, ERGHEYNST P, FROSSARD P. Tree-based pursuit: algorithm and properties[J]. IEEE Trans on Signal Processing, 2006, 54(12): 4685-4697.

【10】MANCERA L, PORTILLA J. L0-norm-based sparse representation through alternate projection[C]. IEEE International Conference on Image Processing, Atlanta. USA, 2006: 2089-2092.

【11】CAI T T, WANG L. Orthogonal matching pursuit for sparse signal recovery with noise[J]. IEEE Transactions on Information Theory, 2011, 57(7): 4680-4688.

【12】SZLAM A, KAVUKCOUGLU K, YANN L C. Convolutional matching pursuit and dictionary training[EB/OL]. [2010-10-3]. http: // arxiv. org / abs/ 1010.0422.

【13】LIAN Qiu-sheng, HAN Dong-mei. Sparse representation by dictionary combined convolutional sparse coding and K-SVD[J]. Systems Engineering and Electronics, 2012, 34(7): 1493-1498.
练秋生, 韩冬梅. 基于卷积稀疏编码和K-SVD联合字典的稀疏表示[J]. 系统工程与电子技术, 2012, 34(7): 1493-1498.

【14】DONG W S, LI X, ZHANG L, et al. Sparsity-based image denosing via dictionary learning and structural clustering[C]. IEEE Conference on Computer Vision and Pattern Recognition, Providence, USA, 2011: 457-464.

【15】XU Jian, CHANG Zhi-guo. Self-adaptive image sparse representation algorithm based on clustering and its application[J]. Acta Photonica Sinica, 2011, 40(2): 316-320.
徐健, 常志国. 基于聚类的自适应图像稀疏表示算法及其应用[J]. 光子学报, 2011, 40(2): 316-320.

【16】SONG Chang-xin, MA Ke, QIN Chuan, et al. Infrared image segmentation based on clutering combined with sparse coding and spatial constraints[J]. Acta Physica Sinica, 2013, 62(4): 1-10.
宋长新, 马克, 秦川, 等. 结合稀疏编码和空间约束的红外图像聚类分割研究[J]. 物理学报, 2013, 62(4): 1-10.

【17】ZHANG Dan-ying, LI Cui-hua, LI Xiong-zong, et al. An image de-noising algorithm based on redundance removed dictionary[J]. Journal of Xiamen University(Natural Science), 2012, 51(4): 691-695.
张丹莹, 李翠华, 李雄宗, 等. 一种基于去冗余字典的图像去噪算法[J]. 厦门大学学报(自然科学版), 2012, 51(4): 691-695.

【18】YE Min-chao, QIAN Yun-tao, SHEN Yan-hao. Clustering based sparse model for image denoising[J]. Journal of Signal Processing, 2011, 27(10): 1593-1598.
叶敏超, 钱沄涛, 沈言浩. 基于聚类的图像稀疏去噪方法[J]. 信号处理, 2011, 27(10): 1593-1598.

【19】WU Ya-dong, SUN Shi-xin. A new hybrid image de-noising algorithm based on 2D wavelet shrinkage and nonlinear diffusion[J]. Acta Electronica Sinica, 2006, 34(1): 163-166.
吴亚东, 孙世新. 基于二维小波收缩与非线性扩散的混合图像去噪算法[J]. 电子学报, 2006, 34(1):163-166.

【20】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.

【21】ENGAN K, SKRETTING K, HUSOY J H. Denoising of images using designed signal dependent frames and matching pursurt[C]. IEEE International Conference on Acoustics, Speech, Signal Processing, Philadelphia, USA, 2005, 653-656.

【22】PU Yun-wei, JIN Wei-dong, ZHU Ming, et al. On cluster validity for kernelized fuzzy C-mean Algorithm[J]. Computer Science, 2007, 34(2): 207-210.
普运伟, 金炜东, 朱明, 等. 核模糊C均值算法的聚类有效性研究[J]. 计算机科学, 2007, 34(2): 207-210.

【23】FRIEDMAN J, HASTIE T, FLING H H, et al. Pathwise coordinate optimization[J]. The Annals of Applied Statistics, 2007, 1(2): 302-332.

引用该论文

WU Yi-quan,LI Li. Image Denoising Using Kernel Fuzzy Clustering and Regularization on Sparse Model[J]. ACTA PHOTONICA SINICA, 2014, 43(3): 0310001

吴一全,李立. 利用核模糊聚类和正则化的图像稀疏去噪[J]. 光子学报, 2014, 43(3): 0310001

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF