提出一种基于纹理的增强型3D矢量场可视化算法,可显著地改善传统纹理法的绘制质量,首先通过对3D纹理的线性卷积运算生成具有空间相关性的卷积纹理;然后对卷积纹理进行高通滤波,以增加流面内流线之间强度的对比;最后通过体绘制方式展示3D卷积纹理,借助权重区域,该算法可以显示用户感兴趣区域或特征区域,避免卷积数据过多引起的紊乱及相互遮挡。
This paper presents an enhanced 3D flow visualization algorithm based on the 3D line integral convolution (LIC) and volume rendering techniques. Given an input 3D vector field, a 3D line integral convolution is at first performed to get a spatial-correlated texture. The resulting texture is then filtered with a 1D high-pass filter, resulting in high intensity contrast among the streamlines in the stream surfaces. An importance-driven volume rendering algorithm is employed to depict the interesting regions, and by which the distracting occlusions and visual clutter are effectively eliminated.