摘要
针对锥束CT图像重建系统中GPU型号不一致问题,提出了基于异构多GPU的重建模型。该模型基于FDK算法进行重建,采用了按计算能力进行任务分配的方法,确保各GPU计算平衡。采用数据流分解的方法,实现了海量数据的图像重建。给出了该重建模型基于CUDA的实现方法,包括采用流管理和异步函数来实现多GPU并行计算以及滤波和反投影核函数的流程设计。利用高精度工业CT系统进行模型的实验验证。结果表明:所建立的重建模型正确有效,能充分发挥系统中异构多GPU的计算能力,执行效率高。
With respect to the problem due to the different multiGPU types in cone beam CT reconstruction, a model was proposed based on heterogeneous multiGPUs. Using FDK algorithm for reconstruction, the model allocated tasks according to the computing capacity of each GPU, ensuring the balance among GPUs. Massive data image reconstruction was achieved by data flow decomposition. The implementation of the method was carried out based on CUDA, including multiGPUs parallel computing using data flow management and asynchronous function and the design of the kernel function in filtering and backprojection. The model was tested on the high precision in dustrial CT system. The results illustrate that the reconstruction model is accurate and effective, taking full advantage of heterogeneous multiGPUs, and is considerably effec tive compared to conventional methods.
出处
《原子能科学技术》
EI
CAS
CSCD
北大核心
2013年第11期2161-2165,共5页
Atomic Energy Science and Technology
关键词
异构多GPU
锥束CT
图像重建
heterogeneous multi-GPU
cone-beam CT
image reconstruction
