关于不适定积分方程修正OS-EM算法的收敛性研究  

Study on convergence of modified OS-EM algorithm in ill-posed integral equations

作　　者：郭瑶 王金平 GUO Yao;WANG Jinping(School of Mathematics and Statistics,Ningbo University,Ningbo 315211,China)

机构地区：[1]宁波大学数学与统计学院

出　　处：《宁波大学学报（理工版）》2019年第6期77-81,共5页Journal of Ningbo University：Natural Science and Engineering Edition

基　　金：国家自然科学基金（61271398）

摘　　要：主要针对修正OS-EM(Ordered-Subset Expectation-Maximization)重建算法进行研究,即利用超松弛参数来加速有序子集,期望最大化的快速重建算法,并且通过OS-EM算法来进行收敛性分析.此外,还充分利用KL距离的一些性质,以探究在精确数据的情况下,修正OS-EM算法的单调性及其方程解的收敛性.In this paper,we mainly study the modified OS-EM(ordered-subset expectation-maximization)reconstruction algorithm,which is a fast reconstruction algorithm that uses overrelaxation parameters to accelerate the expectation of ordered subsets,and analyze convergence by OS-EM algorithm.By making full use of some properties of KL distance,we explore the monotonicity of the modified OS-EM algorithm and the convergence of the solution of the equations in the case of accurate data.

关 键 词：修正OS-EM算法 Kullback-Leibler距离 松弛算子 

分 类 号：O177.6[理学—数学] O175.5[理学—基础数学]