上海交通大学学报

上海交通大学学报（自然版）

• 自动化技术、计算机技术 • 上一篇    下一篇

基于图像处理器的EWA Splatting体绘制加速算法

秦红星，杨杰，郭律，唐炜   

  1. （上海交通大学 图像处理与模式识别研究所， 上海 200240）
  • 收稿日期:2007-12-10 修回日期:1900-01-01 出版日期:2008-12-28 发布日期:2008-12-28

A Hard Accelerated Algorithm for EWA Splatting Based on Modern GPU

QIN Hong-xing,YANG Jie,GUO Lü, TANG Wei   

  1. (Institute of Image Processing & Pattern Recognition, Shanghai Jiaotong University,
    Shanghai 200240, China)
  • Received:2007-12-10 Revised:1900-01-01 Online:2008-12-28 Published:2008-12-28

摘要： 提出了利用目前的图像处理器（GPU）的点块纹理、帧缓存对象（FBO）和可编程功能来加速实现EWA Splatting的体绘制算法.该算法直接以采样点作为代理几何，利用光线投射在GPU中求取每个像素点的所对应三维纹理坐标，同时用一个近似低通滤波器代替高斯滤波器.试验结果表明，所提出的算法可以应用于高质量的交互科学计算可视化.

关键词: 体绘制, 硬件加速, 点块纹理, 图像处理器

Abstract: This paper proposed an accelerated algorithm for EWA volume Splatting based on modern graphic processing unit (GPU). It mainly made use of point sprite, frame buffer object and programmable function of GPU. In the algorithm, sampling points are directly regarded as proxy geometry, three dimension texture coordinates at pixels are computed by ray tracing in GPU and Gaussian filter is replaced by an approximate filter.The experimental results demonstrate that the algorithm can be used in high quality and interactive visualization.

中图分类号: