摘要
电阻抗成像技术是继CT、核磁共振成像后的一个新兴无损伤技术。总结研究了电阻抗成像图像重建的典型物理问题及其求解算法:Levenberg-Marpuardt(LM)迭代方法与负梯度算法,分析了LM迭代方法以及负梯度算法的各自优缺点,将LM迭代和负梯度算法相结合以解决该图像重建问题中的不适定问题(在EIT计算中前期可以用负梯度法迭代,提高收敛速度,后期使用LM算法,保证收敛的精准),并使用了常用的Matlab软件进行数据仿真试验。从试验的重建图像分析来看,采用基于负梯度算法的LM迭代算法,重建图像的精度优于负梯度法和LM迭代法的单个重建结果。
Electrical impedance imaging technology is a newly raised non-invasive technology after Computed Tomography(CT)and Magnetic Resonance Imaging(MRI)technologies.The classical physics problem is concluded and researched in the article,two algorithms such as LM iteration and gradient descent are discussed.Both advantages and disadvantages are analyzed in it,then combined gradient descent with LM iteration,the ill-posed problem is solved.(at the beginning of EIT computation,gradient descent is be used to accelerate convergence.AT the late stage,LM iteration is used,which can make sure of the accuracy of convergence.)Then,the Matlab computer software is used for data simulation.And seeing from the pictures of the experiment,it can be found that the algorithm combined L.M iteration algorithm with gradient descent algorithm is better than single gradient descent algorithm and single LM iteration algorithm.
出处
《长江大学学报（自科版）（上旬）》
CAS
2014年第12期52-54,1,共4页
JOURNAL OF YANGTZE UNIVERSITY (NATURAL SCIENCE EDITION) SCI ＆ ENG
关键词
电阻抗成像
图像重建
反问题
LM算法
负梯度算法
electrical impedance imaging
image reconstruction
inversed problem
LM iteration algorithm
gradient descent
