基于组合字典的图像复原约束优化算法

doi:10.3969/j.issn.1000-3428.2012.21.055

计算机工程 ›› 2012, Vol. 38 ›› Issue (21): 206-209,213.doi: 10.3969/j.issn.1000-3428.2012.21.055

基于组合字典的图像复原约束优化算法

肖宿¹，韩国强²，肖建于¹

(1. 淮北师范大学计算机科学与技术学院，安徽淮北 235000；2. 华南理工大学计算机科学与工程学院，广州 510006)

收稿日期:2012-01-30 出版日期:2012-11-05 发布日期:2012-11-02
作者简介:肖宿(1982－)，男，讲师、博士，主研方向：图像处理；韩国强，教授、博士、博士生导师；肖建于，副教授、博士
基金项目:
国家自然科学基金资助项目(61070090)；国家自然科学基金青年科学基金资助项目(61102117)

Constrained Optimization Algorithm for Image Restoration Based on Combined Dictionaries

XIAO Su ¹, HAN Guo-qiang ², XIAO Jian-yu ¹

(1. School of Computer Science and Technology, Huaibei Normal University, Huaibei 235000, China; 2. School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China)

Received:2012-01-30 Online:2012-11-05 Published:2012-11-02

摘要/Abstract

摘要： 提出一种基于组合字典和约束优化的图像复原算法。建立表示图像复原问题的约束优化模型，其目标函数由l2保真项和双l1正则项的线性组合构成。利用交替优化技术将模型分解为多个子问题求解，并通过邻近算子解决降噪子问题。实验结果表明，与Oliverira算法和Beck算法相比，该算法的复原速度较快，所得图像质量较好，且复原图像与原始图像的均方误差较小。

关键词: 图像复原, 约束优化模型, 稀疏表示, 交替最小化方法, 邻近算子, 软阈值函数

Abstract: According to the existing problems of image restoration in the speed and the quality, this paper presents an image restoration algorithm based on combined dictionaries and constrained optimization. A new constrained optimization model representing the image restoration problem is created. The objective function of it is composed of the l2 data fidelity term and dual l1 regularized term. Then the alternating minimization technology is used to decompose the constrained optimization model into several subproblems to be solved, and the proximal operators are introduced to solve denoising subproblems. Experimental results demonstrate the effectiveness of the presented image restoration algorithm, and compared with some similar state-of-the-art algorithms, it shows better restored results and faster speed.

Key words: image restoration, constrained optimization model, sparse representation, alternating minimization method, proximal operator, soft-thresholding function

中图分类号:

TN911.73

肖宿, 韩国强, 肖建于. 基于组合字典的图像复原约束优化算法[J]. 计算机工程, 2012, 38(21): 206-209,213.

XIAO Qi, HAN Guo-Jiang, XIAO Jian-Xu. Constrained Optimization Algorithm for Image Restoration Based on Combined Dictionaries[J]. Computer Engineering, 2012, 38(21): 206-209,213.

参考文献

[1] Tzikas D G, Likas A C, Galatsanos N P. Variational Bayesian Sparse Kernel-based Blind Image Deconvolution with Student’s-t Priors[J]. IEEE Transactions on Image Processing, 2009, 18(4): 753-764.
[2] Oliveira J P, Bioucas-dias M J, Figueiredo M A T. Adaptive Total Variation Image Deblurring: A Majorization-minimization Approach[J]. Signal Processing, 2009, 89(9): 1683-1693.
[3] Vonesch C, Unser M. A Fast Multilevel Algorithm for Wavelet- regularized Image Restoration[J]. IEEE Transactions on Image Processing, 2009, 18(3): 509-523.
[4] Beck A, Teboulle M. A Fast Iterative Shrinkage-thresholding Algorithm for Linear Inverse Problems[J]. SIAM Journal on Imaging Sciences, 2009, 2(1): 183-202.
[5] Fadili J M, Peyre G. Total Variation Projection with First Order Schemes[J]. IEEE Transactions on Image Processing, 2011, 20(3): 657-669.
[6] Blumensath T, Davies M E. Iterative Thresholding of Sparse Approximations[J]. Journal of Fourier Analysis and Applications, 2008, 14(5-6): 629-654.
[7] 刘晨华, 冯象初, 张力娜. 基于离散小波阈值的偏微分图像去噪[J]. 计算机工程, 2008, 34(15): 196-198.
[8] Setzer S. Operator Splittings, Bregman Methods and Frame Shrinkage in Image Processing[J]. International Journal of Computer Vision, 2011, 92(3): 265-280.
[9] Goldstein T, Osher S. The Split Bregman Method for L1- regularized Problems[J]. SIAM Journal on Imaging Sciences, 2009, 2(2): 323-343.
[10] Steidl G, Teuber T. Removing Multiplicative Noise by Douglas- Rachford Splitting Methods[J]. Journal of Mathematical Imaging and Vision, 2010, 36(2): 168-184.
[11] Figueiredo M A T, Bioucas-Dias J M. Restoration of Poissonian Images Using Alternating Direction Optimization[J]. IEEE Transactions on Image Processing, 2010, 19(12): 3133-3145.
[12] 石光明, 刘丹华, 高大化, 等. 压缩感知理论及其研究进展[J]. 电子学报, 2009, 37(5): 1070-1081.
[13] Elad M, Yavneh I. A Plurality of Sparse Representations is Better than the Sparsest One Alone[J]. IEEE Transactions on Information Theory, 2009, 55(10): 4701-4714.
[14] 李宏, 王宇平, 焦永昌. 解非线性两层规划问题的新的遗传算法及全局收敛性[J]. 系统工程理论与实践, 2005, 25(3): 62-71.
[15] 孙文娟, 李忠范, 王彩玲, 等. 同伦方法求解无约束非凸优化问题的局部极小[J]. 东北师大学报: 自然科学版, 2009, 41(3): 17-20.
[16] Robini M C, Lachal A, Magnin I E. On Total Variation Minimization and Surface Evolution Using Parametric Maximum Flows[J]. International Journal of Computer Vision, 2009, 84(3): 288-307.
[17] O’Sullivan J A, Benac J. Alternating Minimization Algorithms for Transmission Tomography[J]. IEEE Transactions on Medical Imaging, 2007, 26(3): 283-297.
[18] Niesen U, Shah D, Wornell G W. Adaptive Alternating Minimization Algorithms[J]. IEEE Transactions on Information Theory, 2009, 55(3): 1423-1429.
[19] Mallat S. A Wavelet Tour of Signal Processing: the Sparse Way[M]. New York, USA: Academic Press, 2008.
[20] Bauschke H H, Burachik R S, Combettes P L, et al. Fixed-point Algorithms for Inverse Problems in Science and Engineering[M]. Berlin, Germany: Springer, 2011.
[21] Dupe F X, Fadili J M, Starck J L. A Proximal Iteration for Deconvolving Poisson Noisy Images Using Sparse Representations[J]. IEEE Transactions on Image Processing, 2009, 18(2): 310-321.
[22] Selesnick I W, Baraniuk R G, Kingsbury N G. The Dual-tree Complex Wavelet Transform[J]. IEEE Signal Processing Magazine, 2005, 22(6): 123-151.
[23] Charles C, Rasson J P. Wavelet Denoising of Poisson-distributed Data and Applications[J]. Computational Statistics & Data Analysis, 2003, 43(2): 139-148.

基于组合字典的图像复原约束优化算法

Constrained Optimization Algorithm for Image Restoration Based on Combined Dictionaries

PDF

可视化

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 10

[1]	端木春江,左德遥. 锚点领域回归与稀疏表示的图像超分辨率方法[J]. 计算机工程, 2019, 45(5): 194-198.
[2]	徐志刚, 马强, 朱红蕾, 张墨逸. 基于色彩约束与非局部稀疏表示的彩色图像超分辨率重建[J]. 计算机工程, 2019, 45(10): 272-276,282.
[3]	卢佳,保文星. 基于独立空谱残差融合的联合稀疏表示高光谱图像分类[J]. 计算机工程, 2019, 45(1): 246-252.
[4]	张万绪,史剑雄,陈晓璇,汪霖,赵明,周延,牛进平. 基于稀疏表示与引导滤波的图像超分辨率重建[J]. 计算机工程, 2018, 44(9): 212-217.
[5]	刘仲,李立春,李慧启. 一种稀疏度自适应的稀疏傅里叶变换算法 [J]. 计算机工程, 2018, 44(2): 141-146.
[6]	李虎,徐岩. 基于DSKSVD字典学习的语音信号欠定盲源分离算法[J]. 计算机工程, 2018, 44(10): 252-257.
[7]	马海菲,张灿龙,李志欣. 基于L1-APG的红外与可见光目标实时融合跟踪[J]. 计算机工程, 2017, 43(7): 274-280,287.
[8]	魏明俊,许道云,徐梦珂. 于均方差度量分块的自动加权稀疏表示算法[J]. 计算机工程, 2017, 43(5): 174-178,184.
[9]	陈鹏旭,林茂松,梁艳阳,刘宏伟. 基于编码稀疏表示的柔索机器人监测图像去噪算法[J]. 计算机工程, 2017, 43(3): 261-265,270.
[10]	张志武,荆晓远,吴飞. 面向分类的增量字典学习算法[J]. 计算机工程, 2017, 43(10): 167-171,185.
[11]	冯思宇,雷印杰,周新志. 针对单样本人脸的三维部分人脸识别[J]. 计算机工程, 2016, 42(9): 144-150.
[12]	黄丹丹,孙怡. 基于稀疏加权模型的局部判别跟踪[J]. 计算机工程, 2016, 42(9): 226-234.
[13]	张建明,李沛,李旭东,吴宏林. 结合中值滤波与稀疏表示的混合去噪算法[J]. 计算机工程, 2016, 42(9): 240-245.
[14]	侯跃恩,李伟光. 基于稀疏表示的残差连续目标跟踪算法[J]. 计算机工程, 2016, 42(9): 257-261,267.
[15]	陈绪君,朱宇芳,胡君红,马得宇. 滑动窗近似线性依赖稀疏的核递推最小二乘算法[J]. 计算机工程, 2016, 42(8): 64-68.