低秩和稀疏联合重建模型的加速算法  

Acceleration Method for the Combined Models of Low-rank and Sparsity

作　　者：黄进红[1] 周根娇[2] HUANG Jinhong;ZHOU Genjiao(School of Mathematics and Computer Science,Gannan Normal University,Ganzhou 341000,China;School of Science and Technology,Gannan Normal University,Ganzhou 341000,China)

机构地区：[1]赣南师范大学数学与计算机科学学院,江西赣州341000 [2]赣南师范大学科技学院,江西赣州341000

出　　处：《赣南师范大学学报》2018年第6期21-24,共4页Journal of Gannan Normal University

基　　金：江西省教育厅科技项目(160963);赣南师范大学科研基地项目(18zb04).

摘　　要：低秩和稀疏联合重建模型是近些年发展起来的处理信号重建或信号恢复问题较为有效的方法之一,诸多研究者的实验表明它比单独使用低秩或稀疏约束更为有效和鲁棒,但传统的求解算法存在速度较慢的问题,这在一定程度上限制了模型的应用.文章利用传统方法的加速版本,对磁共振图像重建中涉及的低秩和稀疏联合重建模型提出了一种新的快速求解算法,并利用动态磁共振图像重建的仿真实验说明了算法的有效性.Low rank and sparse joint reconstruction model developed in recent years,is one of the effective methods to process the signal reconstruction or restoration problems.Many researchers’results show that it is more effective than the single use of low rank or sparse constraint and robust,but its slower speed inherent in the traditional algorithm restricts the application to a certain extent.With the accelerated version of the traditional method,a new fast algorithm for low rank and sparse joint reconstruction model involved in MRI image reconstruction is proposed,and the effectiveness of the algorithm is demonstrated by the simulation experiment of dynamic MRI image reconstruction.

关 键 词：稀疏 低秩 磁共振图像重建 临近梯度算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]