复杂三维模型的遮挡检测加速算法

在环境光照下渲染复杂三维植被模型时,其自身阴影的计算极为耗时;对它进行预处理是加速运算的一个重要方法。笔者对三维模型各处的自身阴影遮挡测试的预处理计算进行优化,即按坐标轴向对三维模型进行分层,用每层的正交投影图像代替层内的复杂三维结构,使遮挡计算可以快速进行;同时还利用层层合并的技术以节省存储空间。实验结果表明,笔者提出的算法明显提高了三维模型自身阴影遮挡测试的运算速度。

An Accelarated Algorithm for Visibility-Testing of Complicated 3D Model

The calculation of self-shadow is very time-consuming while rendering a complicated 3-D plant model with environmental lighting. And an important way dealing with it is pre-computation. In this paper we accelerate visibility-testing process for pre-computing self-shadow of 3-D models. 3-D models are split into thin levels along coordinate axes, and all polygons in this level are projected orthogonally on a plane in the middle of the level as one sheet in order to origin complex models. At the same time, the sheets far from the viewpoints are merged in order to save memory space. The results show that the proposed algorithm is efficient for self-shadow visibility test of complex 3-D model objects, such as tree models.