| 锥束CT重建FDK算法的两级并行计算研究 |
| |
| 引用本文： | 江鹏,陈志强,邢宇翔.锥束CT重建FDK算法的两级并行计算研究[J].核电子学与探测技术,2006,26(1):87-90. |
| |
| 作者姓名： | 江鹏 陈志强 邢宇翔 |
| |
| 作者单位： | 清华大学工程物理系,北京,100084;清华大学工程物理系,北京,100084;清华大学工程物理系,北京,100084 |
| |
| 摘 要： | 三维锥束CT图像的FDK算法重建由于运算量大,在重建高分辨率的图像时,重建所需时间通常难以满足实际应用的需求,集群并行计算是解决上述问题的常用方法之一.在一个SMP集群系统上,采用MPI和Pthreads两种模型相结合的方法,通过节点之间的消息传递和节点内部的共享内存,实现了FDK算法的两级并行.
|
| 关 键 词： | 图像重建 FDK算法 MPI Pthreads 并行计算 |
| 文章编号： | 0258-0934(2006)01-0087-04 |
| 修稿时间： | 2004年12月3日 |
Two-level parallelism of cone-beam CT image reconstruction using the FDK algorithm |
| |
| JIANG Peng,CHEN Zhi-qiang,XING Yu-xiang.Two-level parallelism of cone-beam CT image reconstruction using the FDK algorithm[J].Nuclear Electronics & Detection Technology,2006,26(1):87-90. |
| |
| Authors: | JIANG Peng CHEN Zhi-qiang XING Yu-xiang |
| |
| Abstract: | Image reconstruction of 3D cone-beam tomography using the FDK algorithm requires huge computational resource.The reconstruction of a high resolution 3D object would usually take a long time,which is intolerable in real applications.Parallel computing using cluster systems is one of the common approaches to solve this problem.The mixed model with both MPI and Pthreads is used to parallelize the FDK algorithm on a SMP cluster.The computation is parallelized in a two-level manner with both Message-Passing among nodes and Shared-Memory within each node.Our simulation results on the SMP cluster system are reported. |
| |
| Keywords: | image reconstruction FDK algorithm MPI Pthreads parallel computation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
