基于两相流电容层析成像系统的正则化Landweber算法  

Regularized Landweber Algorithm Based on Two-Phase Flow Electrical Capacitance Tomography System

作　　者：田沛[1] 秦京建 秦旭刚 Tian Pei;Qin Jingjian;Qin Xugang(Department of Automation,North China Electric Power University,Hebei,Baoding,071003,China)

机构地区：[1]华北电力大学自动化系,河北保定071003

出　　处：《仪器仪表用户》2018年第9期90-92,共3页Instrumentation

摘　　要：针对电容层析成像系统的不适定性问题,提出了一种正则化Landweber阈值迭代改进算法。该算法将正则化项引入初值等式约束中,得到带约束的正则化问题,并在Landweber迭代算法的过程中设置门限阈值,在每次迭代中都自适应的调节阈值参数,最终用带约束的正则化Landweber阈值迭代算法得到最终的重建图像结果。数值仿真实验表明,正则化Landweber阈值迭代算法在改善图像重建质量和提高收敛速度方面均有明显的优势。Aiming at the ill posed problem of electrical capacitance tomography(ECT)system,a regularized iterative algorithm for Landweber threshold is proposed.In the algorithm,the regularization term is introduced into the constraint of initial value inequality,and the regularization problem with constraint is obtained.The threshold threshold is set in the process of Landweber iterative algorithm.The threshold parameters are adjusted adaptively in each iteration.Finally,the final reconstructed image results are obtained by the regularized Landweber projection iteration algorithm with constrained regularization.Numerical simulation experiments show that regularization Landweber threshold iterative algorithm has great advantages in improving image reconstruction quality and improving convergence speed.

关 键 词：电容层析成像 Landweber 门限阈值 正则化 图像重建 

分 类 号：TB9[一般工业技术—计量学] TP391.41[机械工程—测试计量技术及仪器]