北京生物医学工程

基于CUDA的快速光线投射法

Accelerated Ray Casting Algorithm Based on CUDA

作者：董现玲江贵平张煜

单位：南方医科大学生物医学工程学院（广州510515）

分类号：

出版年·卷·期（页码）：2010·29·2（125-129）

摘要：

NVIDIA的GeForce 8系列显卡上的统一设备架构（common unified device architecture，CUDA）不需要映射到图形API便可在GPU上进行计算的分配和管理，因此较适合于大数据量的医学图像可视化领域。本文分析了CUDA的设计思想和编程模式，针对其特点对传统基于硬件的光线投射法进行改进，将计算耗时的绘制部分改造成单指令多数据模式（single instruction multiple data，SIMD），并分别运用纹理存储器和共享存储器对算法进行优化。实验结果表明,该方法能够快速、高效地生成可视化图像。

The common unified device architecture （CUDA） on NVIDZA′S GeForce 8 series graphics cards is a fundamentally new programming approach making use of the unified shader design of most current graphics processing units （GPU），therefore it is more suitable for large data visualization of medical images. In this paper we analyzed the design ideas and programming models of CUDA，and then the new technology was applied to volume ray casting algorithm in the field of medical image processing. The normal GPUbased ray casting algorithm was improved and rendering process was redesigned into single instruction multiple data （SIMD） mode. Finally suggestions were presented on how to optimize the algorithm by means of texture memory and shared memory. The experiment results showed that it could achieve high quality image and interactive frame rates by using CUDA.

参考文献：

［1］Levoy M. Display of Surfaces from Volume Data［J］. IEEE Computer Graphics and Applications， 1988， 8（3）： 29-37.
［2］Cullip J， Neumann U. Accelerated Volume Reconstruction with 3D Texture Hardware，TR93-0027［R］. USA： Chapel Hill，NC，1994.
［3］Cabral B， Cam N， Foran J. Accelerated volume rendering and tomographic reconstruction using texture mapping hardware［J］. Proceedings of IEEE Symposium on Volume Visualization， 1994： 91-98.
［4］Krüger J， Westermann R. Acceleration techniques for GPU-based volume rendering［C］//Proceedings of the 14th IEEE Visualization 2003. Washington DC， USA： IEEE Computer Society，2003.
［5］Roettger S， Guthe S， Weiskopf D， et al. Smart hardware-accelerated volume rendering［C］//Proceedings of the Symposium on Data Visualization 2003. Switzerland： Eurograghics Association，2003.
［6］Stegmaier S， Strengert M， Klein T， et al. A simple and flexible volume rendering framework for graphics-hardware-based raycasting［C］//Proceedings of Volume Graphics 2005. Stony Brook， N Y， USA： Eurographics Association， 2005.
［7］NVDIA CUDA programming guide 2.0 beta. NVIDIA， 2008.
［8］Marsálek，L， Hauber A，Slusallek P. High-speed volume ray casting with CUDA ［J］. IEEE Symposium on Interactive Ray Tracing， 2008： 185.
［9］Amy Williams， Steve Barrus，Morley RK， et al. An Efficient and Robust Ray-Box Intersection Algorithm ［J］. Journal of Graphics Tools， 2005 10：54.
［10］唐泽圣. 三维数据场可视化［M］. 北京：清华大学出版社， 1999：33-34.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录