文献类型：期刊文献

中文题名：Learning block-structured incoherent dictionaries for sparse representation

英文题名：Learning block-structured incoherent dictionaries for sparse representation

作者：Zhang YongQin[1];Xiao JinSheng[2];Li ShuHong[3];Shi CaiYun[4];Xie Guoxi[4]

第一作者：Zhang YongQin

通讯作者：Xiao, JS[1]

机构：[1]Northwest Univ, Sch Informat Sci & Technol, Xian 710127, Peoples R China;[2]Wuhan Univ, Sch Elect Informat, Wuhan 430079, Peoples R China;[3]Henan Univ Econ & Law, Coll Comp & Informat Engn, Zhengzhou 450002, Peoples R China;[4]Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen Key Lab MRI, Shenzhen 518055, Peoples R China

第一机构：Northwest Univ, Sch Informat Sci & Technol, Xian 710127, Peoples R China

通讯机构：[1]corresponding author), Wuhan Univ, Sch Elect Informat, Wuhan 430079, Peoples R China.

年份：2015

卷号：0

期号：10

起止页码：1-15

中文期刊名：中国科学：信息科学（英文版）

外文期刊名：SCIENCE CHINA-INFORMATION SCIENCES

收录：;EI(收录号：20152200883139);Scopus(收录号：2-s2.0-84946496003);WOS:【SCI-EXPANDED(收录号:WOS:000362698900007)】；CSCD:【CSCD2015_2016】；

基金：This work was supported by National Natural Science Foundation of China (Grant Nos. 61201442, 61471272), and China Postdoctoral Science Foundation (Grant No. 2013M530481).

语种：英文

中文关键词：学习算法;稀疏表示;块结构;非相干;字典;图像处理;图像信号;聚类方法

外文关键词：dictionary learning; sparse representation; sparse coding; block sparsity; mutual coherence

摘要：Dictionary learning is still a challenging problem in signal and image processing. In this paper, we propose an efficient block-structured incoherent dictionary learning algorithm for sparse representations of image signals. The constrained minimization of dictionary learning is achieved by iteratively alternating between sparse coding and dictionary update. Without relying on any prior knowledge of the group structure for the input data, we develop a two-stage clustering method that identifies the underlying block structure of the dictionary under certain restricted constraints. The two-stage clustering method mainly consists of affinity propagation and agglomerative hierarchical clustering. To meet the conditions of both the upper bound and the lower bound of the mutual coherence of dictionary atoms, we introduce a regularization term for the objective function to adjust the block coherence of the overcomplete dictionary. The experiments on synthetic data and real images demonstrate that the proposed dictionary learning algorithm has lower representation error, higher visual quality and better reconstructed results than most of the state-of-the-art methods.
Dictionary learning is still a challenging problem in signal and image processing. In this paper, we propose an efficient block-structured incoherent dictionary learning algorithm for sparse representations of image signals. The constrained minimization of dictionary learning is achieved by iteratively alternating between sparse coding and dictionary update. Without relying on any prior knowledge of the group structure for the input data, we develop a two-stage clustering method that identifies the underlying block structure of the dictionary under certain restricted constraints. The two-stage clustering method mainly consists of affinity propagation and agglomerative hierarchical clustering. To meet the conditions of both the upper bound and the lower bound of the mutual coherence of dictionary atoms, we introduce a regularization term for the objective function to adjust the block coherence of the overcomplete dictionary. The experiments on synthetic data and real images demonstrate that the proposed dictionary learning algorithm has lower representation error, higher visual quality and better reconstructed results than most of the state-of-the-art methods.