论文部分内容阅读
对现有的电容成像图像重建算法进行了综述并介绍了其原理,其中包括线性反投影法(LBP)、基于奇异值分解(SVD)的直接算法、Tikhonov正则化方法、Newton-Raphson算法、最速下降法、Landweber迭代算法、代数重建技术(ART)、同步迭代重建技术(SIRT)和基于模型的重建算法。在此基础上使用仿真和实验数据对目前主要使用的线性反投影法、基于奇异值分解的直接方法、Tikhonov正则化方法、Tikhonov迭代、投影Landweber迭代等进行了评价。对它们从电容值误差、图像误差、相关系数和耗用时间几个角度进行了比较,并对今后电容成像图像重建算法的发展方向提出了一些看法。
The existing capacitive imaging reconstruction algorithms are summarized and their principles are introduced, including linear backprojection (LBP), direct algorithm based on singular value decomposition (SVD), Tikhonov regularization method, Newton-Raphson algorithm, Descent method, Landweber iterative algorithm, algebraic reconstruction technique (ART), synchronous iterative reconstruction technique (SIRT) and model-based reconstruction algorithm. On this basis, we use the simulation and experimental data to evaluate the main linear backprojection method, direct method based on singular value decomposition, Tikhonov regularization method, Tikhonov iteration and projection Landweber iteration. They are compared from the perspectives of capacitance error, image error, correlation coefficient and elapsed time, and some suggestions are put forward for the future development of capacitance imaging algorithm.