期刊文献+

GPU多重纹理加速三维CT图像重建 被引量:8

Fast 3D CT reconstruction with multitexture of GPU.
下载PDF
导出
摘要 三维锥束CT图像重建运算量大,纯软件(仅使用CPU)计算时间较长。为了充分利用计算机图形处理器(Graphic Process Unit,GPU)的并行处理能力以及提高数据传输效率,研究了一种结合使用GPU多重纹理(multitexture)加速三维锥束CT的FDK图像重建过程的方法。该方法采用多重纹理映射来提高反投影速度、减少中间数据存储量、减少浮点累加次数,使用顶点颜色通道来实现距离加权运算,采用扩展方法来增加并行反投影的纹理单元,从而提高重建速度。计算机实验结果表明,使用普通PC机重建尺寸为2563的图像,在保证数据精度为16bit浮点数的要求下,GPU反投影计算可以在10s以内完成。与仅使用CPU的重建方法相比,GPU重建图像加速方法达到了较高的时间加速比。 The three-dimension Computed Tomography(CT) reconstruction is time consuming on CPU only.A new accelerated 3D reconstruction method with FDK algorithm using multitexture of GPU (Graphic Process Unit) was developed.Uslng muhitexture hardware,the reconstruction speeds up the baekprojeetion,reduee the temporary memory for saving the result of backprojeetion, and decrease the floating-point sum passes.Using the color channels of vertexes to realize FDK depth weighting computation,and extending the texture units for backprojection,the reconstruction is accelerated.Therefore,the backprojection of a 2563 volume can be completed with 16 bit floating-point textures in less than 10 seconds.The experiment results show that the new GPU accelerated 3D reconstruction method get a fast performance and good quality.
作者 马车平 曾理
出处 《计算机工程与应用》 CSCD 北大核心 2008年第7期227-230,共4页 Computer Engineering and Applications
基金 国家自然科学基金(the National Natural Science Foundation of China under Grant No.600672098)
关键词 三维锥束CT FDK算法 加速重建 图形处理器 多重纹理 3D cone CT FDK algorithm accelerated reconstruction GPU(Graphic Process Unit) multitexture
  • 相关文献

参考文献10

  • 1Kak A C,Slaney M.Principles of computerized tomographic imaging[M}.New York:IEEE Press, 1988.
  • 2Feldkamp L A,Davis L C,Kress J W.Practical cone-beam algorithm[J].Optical Society of America, 1984,1 (6):612-619.
  • 3Xu Fang,Mueller K.Aceelerating popular tomographic reconstruction algorithms on commodity PC graphics hardware[J].IEEE Transaction on Nuclear Science, 2005,52(3 ) : 654-663.
  • 4Xu Fang,Mueller K.Towards a unified framework for rapid 3D computed tomography on commodity GPUs[J].IEEE Nuclear Science Symposium Conference Record,2003,4: 2757-2759.
  • 5Xu Fang,Mueller K. Ultra -fast 3D filtered backprojection on commodity graphics hardware[J].IEEE International Symposium on Biomedical Imaging: Macro to Nano,2004, 1 : 571-574.
  • 6Mueller K,Yagel R,Rapid 3-D cone-beam reconstruction with the Simultaneous Algebraic Reconstruction Technique(SART) Using 2-D texture mapping hardware[J].IEEE Trans Med Imag,2000,19 ( 12 ) : 1227-1237.
  • 7Mueller K, Xu F.Practical considerations for GPU-accelerated CT[J]. IEEE Intermational Symposium on Biomedical Imaging:Macro to Nano, 2006 : 1184-1187.
  • 8Kapasi U,Dally W,Khailany B,et al.Programmable stream processors[J].IEEE computer, 2003,36( 8 ) : 54-62.
  • 9梁亮,张定华,毛海鹏,顾娟.一种基于可编程图形硬件的快速三维图像重建算法[J].计算机应用研究,2006,23(1):241-243. 被引量:5
  • 10戴智晟,陈志强,邢宇翔,张丽,毕文元.用通用显卡加速三维锥束T-FDK重建算法[J].清华大学学报(自然科学版),2006,46(9):1589-1592. 被引量:6

二级参考文献14

  • 1吴恩华,柳有权.基于图形处理器(GPU)的通用计算[J].计算机辅助设计与图形学学报,2004,16(5):601-612. 被引量:225
  • 2Feldkamp LA, Davis LC, Kress JW. Practical Cone-beam Algorithm[J]. J Opt. Soc. Am.A,1984,1(6): 612-619.
  • 3.[EB/OL].http://www.gpgpu.org[EB/OL],.
  • 4Joao Luiz Dihl Comba, Carlos A Dietrich, Christian A Pagot, et al.Computations on GPUs: From a Programmable Pipeline to an Efficient Stream processor [ J ]. Revista de Informatica Teoricae Aplicada,2003, X(2).
  • 5Klaus Mueller, Roni Yagel. Rapid 3-D Cone-beam Reconstruction with Simultaneous Algebraic Reconstruction Technique(SART) Using 2-D Texture Mapping Hardware [ J ]. IEEE Transactions on Medical Imaging, 2000,19(12): 1227-1237.
  • 6M Segal, C Korobkin, R van Widenfelt ,et al. Fast Shadows and Lighting Effects Using Texture Mapping [ C ]. SIGGRAPH '92, vol.26,1992. 249-552.
  • 7Mark W R, Glanville, et al. Cg: A System for Programmable Graphics Hardware in a C-like Language[J]. ACM Transactions on Graphics, 2003, 22(3):806-907.
  • 8Cabral B,Cam N,Foran J.Accelerated volume rendering and tomographic reconstruction using texture mapping hardware[C]∥ Symposium on Volume Visualization.New York:ACM Press,1994:91-98.
  • 9XU F,Mueller K.Accelerating popular tomographic reconstruction algorithms on commodity PC graphics hardware[J].IEEE Trans Nuclear Sci,2005,52(3):654-663.
  • 10XU F,Mueller K.Ultra-fast 3D filtered backprojection on commodity graphics hardware[J].IEEE International Symposium on Biomedical Imaging:Macro to Nano,2004,1:571-574.

共引文献9

同被引文献48

引证文献8

二级引证文献11

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部