基于空间网格细分的不规则场景的光线跟踪

针对不规则场景中光线跟踪算法绘制速度慢的问题，在深入学习和比较近些年的光线跟踪加速算法的基础上，提出了一种改进的网格细分的光线跟踪算法。首先，设置矩形场景包围盒，剔除对场景没有影响的外部光线，进而简化求交运算；其次，采用新方法创建空间网格，该方法可使空间单元数量和存储空间复杂度都限定在一定范围内；最后，对网格进行细分，这一步骤消除了传统空间网格算法忽略部分空白区域对加速效果产生的不良影响, 极大完善了传统空间网格算法。通过实验证明，该方法能有效提高光线在空白空间的穿行速度，不仅提高了时间效率，而且减少了空间开销。

Ray tracing of irregular scene based on spatial grid subdivision

To solve the slow rendering problem of ray tracing algorithm in irregular scene, an improved grid subdivision ray tracing algorithm was proposed based on the in-depth study and comparison of the recent acceleration algorithms of ray tracing. First, the rectangular bounding box of the scene was set to remove the influence of external light, and the intersect operations was simplified; second, spatial grids were created with a new way to limit the spatial unit number and complexity of storage space within a certain range; finally, the traditional spatial grid algorithm was greatly improved by subdividing grids to eliminate the bad influence on acceleration effects due to ignoring some blank space. The experimental results show that this method can effectively improve the light speed in blank space, it not only increases the time efficiency but also reduces the space lost.