离散化代数重建的全变差算法改进

摘要

This paper presents an improved discretization Algebraic Reconstruction algorithm which bases on Total Variation, mainly for discretization Algebraic Reconstruction (DART) algorithm is easily affected by factors such as noise and make the problem of image edge blurred, using Total Variation (TV) to minimize the constraints, an improved DART Reconstruction algorithm.Experiments show that the algorithm is compared with the traditional ART algorithm, can quickly reconstruct image, compared with the DART algorithm, improved the condition of image edge blur and has good noise resistance.%文中提出一种离散化代数重建的全变差改进算法,主要针对离散化的代数重建算法易受到噪声等因素的影响而使图像边缘较为模糊的问题,利用全变差最小化的约束条件,提出一种改进的DART重建算法.实验表明,该算法与传统ART算法相比,能较快重建出图像,与DART算法相比,改善图像边缘模糊的情况,具有较好的抗噪性.