一种基于FDK算法的扁平物体局部重建方法
A Method for the Local Reconstruction of Flat Object Based on FDK Algorithm
-
摘要: 投影数据有截断的局部重建问题是CT重建领域的一个难点,传统的重建算法会产生严重的截断伪影,最近提出的BPF算法和POCS迭代方法虽然能解决局部重建问题,但是它们的重建效率和并行性都比较差。针对特殊的扁平形状物体的局部重建问题,本文对FDK算法在此类问题上的适用性进行了分析研究。通过数字仿真实验,给出了一个可以比较准确重建扁平物体局部区域的条件,即:局部区域在水平方向上的长度大于该物体在水平方向上长度的1/8,物体厚度小于该物体在水平方向上长度的1/13。数字仿真和真实数据的重建结果证实了在这个条件下FDK算法可以很好地实现扁平物体的局部重建。Abstract: The local reconstruction with truncated projection data is a difficult problem in CT reconstruction field.Conventional reconstruction algorithms will product severe truncated artifacts.The rising BPF algorithm and POCS iterative algorithm can solve the local reconstruction problem,but their reconstruction efficiency and paralleling performance are poor.In this paper,aiming at special structural flat object,it is found that if the thickness and the horizontal length of local area satisfy some conditions,accurate reconstruction can be carried out using efficient FDK algorithm.The results of numerical experiments show that when the horizontal length of local region is bigger than 1/8 horizontal length of object and the thickness is less than 1/13 horizontal length of object,the reconstruction results are comparable with the true values.The reconstruction results of real data also confirmed FDK algorithm could reconstruct the local area of flat object well under the condition.