摘要
计算机层析成像技术(Computer Laminography,CL)主要用于大尺寸、低厚度板状物体的无损检测。迭代重建算法虽然适用于CL这种角度受限的扫描方式,但是却非常耗时。本文研究了CL系统SART迭代重建算法的GPU加速方法。介绍了基于线驱动的投影,和基于线驱动及体素驱动的反投影的GPU加速方法,实验结果表明,采用GPU加速可以达到254倍的加速比。
Computed Laminography(CL)is mainly for the non-destructive testing of large-sized,thin objects such as plates.Iterative reconstruction algorithms are applicable to CL,while they are time consuming.The GPU accelerated simultaneous algebraic reconstruction technique(SART)was studied in this work.Ray-driven projection/backprojection and voxel-driven backprojection model,as well as their GPU accelerated version were introduced.The results show that GPU accelearation can reach 58 times speedup.
作者
贾统
刘双全
孙校丽
魏存峰
乔志伟
刘宝东
JIA Tong;LIU Shuangquan;SUN Xiaoli;WEI Cunfeng;QIAO Zhiwei;LIU Baoclong(Beijing Engineering Research Center of Radiographic Techniques and Equipment,Institute of High Energy Physics,Chinese Academy of Sciences,Beijing 100049,China;School of Nuclear Science and Technology,University of Chinese Academy of Sciences,Beijing 100049,China;School of Computer and Information Technology,ShanxiUniversity,Taiyuan Shanxi 030006,China)
出处
《中国体视学与图像分析》
2020年第4期393-400,共8页
Chinese Journal of Stereology and Image Analysis
基金
国家重点研发计划资助(2017YFF0107201)
中国科学院创新交叉团队项目资助(JCTD-2019-02)
山西省重点研发计划(201803D421012)
山西省回国留学人员科技活动择优资助项目(RSC1622)
山西省回国留学人员科研资助项目(2020-008)。
关键词
计算机层析成像
联合代数重建算法
GPU加速
computer laminography
simultaneous algebraic reconstruction technique
GPU acceleration
