稀疏表示及其算法分析  

Sparse representation and its algorithm analysis

作　　者：么炜[1] 刘博[1] 滕桂法[2] 

机构地区：[1]河北农业大学信息科学与技术学院,河北保定071000 [2]河北农业大学研究生学院,河北保定071000

出　　处：《河北农业大学学报》2014年第2期115-119,共5页Journal of Hebei Agricultural University

基　　金：河北省高等学校科学技术研究青年基金项目(QN20131151);保定市科学技术研究与发展计划(12ZG003);河北农业大学非生命学科与新兴学科科研发展基金

摘　　要：利用信号本身的结构特征,通过附加不同的稀疏约束,该模型利用过完备字典进行信号分解,使其表示成字典中若干原子的线性组合,从而获得数据的精简表示。该表示具有自适应性,鲁棒性等特点,使其在信号与图像处理方面有着广泛的应用。本文分析了稀疏表示的常见模型与主要求解算法,并研究了稀疏表示在图像重构与图像分类中的应用。In term of the intrinsic structured characteristic of signal itself, this model decompo- ses a signal as a linear combination of a few atoms from an over-completed dictionary. As it turns out, we obtain the parsimonious representation of signal with regularization by different sparisty-inducing norms. Through the adaptivity and robustness of sparse representation, it is well applied to the field of signal and image processing. In this paper, we analyze the main sparse models and el-minimization approaches, and study its applications in image reconstruc- tion and classification. Finally, main problems of this model are discussed, as well as the fu- ture research.

关 键 词：稀疏表示 稀疏编码 结构化稀疏 字典学习 l1-范数 

分 类 号：TP391.6[自动化与计算机技术—计算机应用技术]