%0 Journal Article
%T 压缩感知中基于广义Jaccard系数的gOMP重构算法<br>gOMP reconstruction algorithm based on generalized Jaccard coefficient for compressed sensing
%A 张晓东
%A 董唯光
%A 汤旻安
%A 郭俊锋
%A 梁金平&lt
%A br&gt
%A ZHANG Xiao-dong
%A DONG Wei-guang
%A TANG Min-an
%A GUO Jun-feng
%A LIANG Jin-ping
%J 山东大学学报（理学版）
%D 2017
%R 10.6040/j.issn.1671-9352.0.2017.093
%X 摘要： 为了解决信号重构性能差的问题,提出了一种基于广义Jaccard系数的广义正交匹配追踪(generalized orthogonal matching pursuit, gOMP)重构算法。该算法利用广义Jaccard系数相似性匹配准则替换gOMP算法中的内积度量准则,优化了通过感知矩阵来选择与残差余量最匹配原子的匹配方式。实验结果表明,该算法的重构成功率不仅高于gOMP算法,同时也高于OMP、StOMP等算法。<br>Abstract: In order to solve the problem such as low reconstruction performance of the signal, we propose a generalized orthogonal matching pursuit(gOMP)reconstruction algorithm based on generalized Jaccard coefficient. The improved gOMP algorithm replaces matching criterion of inner product by similarity matching criterion of generalized Jaccard coefficient, and selects the most matching atom from projection matrix and residual signal. Experimental results show that the reconstruction success rate of proposed algorithm is much better than other algorithms, such as gOMP, OMP, StOMP and so on
%K 压缩感知
%K 广义Jaccard系数
%K 相似性匹配准则
%K 广义正交匹配追踪
%K 重构算法
%K &lt
%K br&gt
%K compressed sensing
%K generalized orthogonal matching pursuit
%K similarity matching criterion
%K reconstruction algorithm
%K generalized Jaccard coefficient
%U http://lxbwk.njournal.sdu.edu.cn/CN/10.6040/j.issn.1671-9352.0.2017.093