基于粒子纹理融合的流场可视化方法

基于纹理的可视化方法可以描述流场的整体结构，但传统方法计算耗时，生成可视化图像对比度比较低。从加速可视化整体流程出发，提出了一种基于粒子纹理融合的流场可视化方法。此方法首先随机产生一组噪声图像作为初始粒子分布图，然后依次将初始粒子分布图与根据流动而变形的数据网格加权融合得到粒子轨迹图，最后一帧帧彼此相邻的粒子轨迹图组成一个流场的动态显示。该方法具有独立于流场数据、绘制速度快、生成图像对比度高的特点，参数物理意义明显，不同参数选择可得到不同视觉效果的可视化输出结果，能够充分利用现有硬件图形显示加速设备,已经被成功应用于空间晶体生长实验流场数据的可视化，获得了较好的效果。

Flow field visualization method based on particle texture blending

Texture based visualization methods can describe the whole structure of the flow field, but traditional methods are always time consuming and the visualization image has low contrast. In order to improve the efficiency of the whole visualization process, a new particle texture blending based flow field visualization method was presented. Firstly a set of random noise images were given as original particle distribution images. Then particle trace image would be generated by blending the original particle distribution image with distorted flow data grid. Finally, a series of particle trace images compose the dynamic display of flow field. This method is independent from flow data, having fast rendering rate and high contrast output image. The parameters have obvious physical meaning, and different vision effect visualization output results are derived from different parameters combination. This method can take full advantage of the graphic card accelerator, and has successful application in the dynamic visualization of the flow data in crystal growth experiment during task research work.