体绘制技术常用于3维体数据场的可视化,其虽然可以生成高质量的投影图像,但通常不能绘制由体数据与点、线、面图形组成的混合场景。在现有的混合场景可视化方法中,有些只能绘制由体数据与面图形组成的复杂混合场景,而不能处理存在点和线的混合场景;有的则成像速度慢、成像质量差。为了能够正确地绘制复杂混合场景,采用SIMD和软件加速等技术,提出了一种速度快、成像质量高的基于光线投射算法的混合场景可视化方法,并分析了该算法所具有的3种绘制次序,以便满足不同应用的要求。该算法既可用于不同场景的绘制,又可用于平行和透视投影中。实验结果表明,该算法能够正确地绘制体数据与点、线、面图形组成的混合场景,且成像速度快,图像质量高。

Volume rendering is often used for 3D volumetric datasets visualization.It can generate high quality images,but can not render volumes and polygons combined scenes.Among many currently available techniques of visualizing hybrid scenes,some can render volumes and triangles combined scenes,but can not process geometric points and lines.To correctly render these complicated hybrid scenes,a ray casting based algorithm using SIMD and software acceleration techniques is presented.The algorithm has three types of rendering order,each of which has its own advantages and drawbacks for different applications.It can be adopted in parallel and perspective projection for many applications.Experiments show that the algorithm can properly visualize volumes,polygons,lines and points combined scenes,and rendered images have high quality and its rendering speed is very fast.