代数重建和同步迭代重建在电容层析成像中的比较研究  

Comparison of Algebraic Reconstruction Technique and Simultaneous Iterative Reconstruction Technique in Electrical Capacitance Tomography Image Reconstruction

作　　者：陈夏[1] 胡红利[1] 高享想[1] 方天驰[1] 

机构地区：[1]西安交通大学电气绝缘与电力设备国家重点实验室,西安710049

出　　处：《西安交通大学学报》2011年第4期25-29,共5页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目(50777049);国家"863计划"资助项目(2009AA04Z130)

摘　　要：针对电容层析成像技术(ECT)中的逆问题——图像重建算法的非线性和病态性问题,以12电极电容层析成像系统为对象,研究了代数重建算法(ART)与同步迭代重建算法(SIRT).分别对仿真电容值和实测电容值进行图像重建的实验验证,从成像效果、相对误差及重建时间3个方面对这两种算法进行了评估与分析.结果表明,ART算法和SIRT算法均能有效地实现ECT图像重建,其中SIRT算法能够在100次迭代内达到高精度,在收敛速度和成像效果上更具有优势.To solve the nonlinear and ill-posed inverse problems in electrical capacitance tomography（ECT）, algebraic reconstruction technique （ART） and simultaneous iterative reconstruction technique （SIRT）were put forward to meet the requirement of image reconstruction. 12-electrode ECT system was set up to acquire real measurement capacitances under annular flow and stratified flow. The imaging results of the two algorithms in simulated capacitances and measured capacitances were evaluated and analyzed respectively for the imaging quality, relative error and reconstruction time. The experiments indicated that ART and SIRT for the presented ECT system perform effectively. Compared with ART, the convergence rate and the quality of SIRT are more advantageous with higher accuracy within 100 iterations.

关 键 词：电容层析成像技术 图像重建 代数重建 同步迭代重建 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]