自适应最小梯度夹角预积分光照算法  

Adaptive Pre-Integrated Illumination Algorithm with Minimum Gradient Angle

作　　者：徐超清 刘义鹏[1] 马祥音[1] 徐冰[1] 梁荣华[1] 

机构地区：[1]浙江工业大学信息工程学院,杭州310023

出　　处：《计算机辅助设计与图形学学报》2016年第1期41-49,共9页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61379076;61202205);浙江省自然科学基金(LQ15F020009)

摘　　要：为了增强体绘制的纹理细节与光照效果,提出了一种自适应最小梯度夹角预积分算法.通过采样点的方向导数及空间位置等信息来确定采样光线上的极值点,利用自适应划分算法加强极值点区域的绘制;为了突出物体的真实感、增强纹理细节,引入预积分光照算法,根据最小梯度夹角算法计算预积分采样段的光照颜色值,最后对预积分采样段进行体绘制积分.实验结果表明,相比传统的预积分算法,文中算法针对标量变化较大的区域能够得到较好的绘制效果,并且能够增加局部细节的光照效果.To enhance volume rendering texture details and optimize the illumination effects, this paper presents a novel approach for adaptive pre-integrated illumination with minimum gradient angle. The extreme points of the sampled light are determined by directional derivative and spatial position. The rendering effect for extreme point region can be enhanced by the adaptive partitioning method. In order to highlight the sense of the object reality and enhance texture details, we introduce pre-integrated illumination algorithm in the paper. The calculation of color value of the pre-integral sampling section is based on minimum gradient angle algorithm. Then we can render the pre-integral sampling part. Experimental results show that, in contrast to the traditional pre-integrated method, our method can get better results where the scalar change greatly, and it can enhance the illumination effect for local details.

关 键 词：直接体绘制 自适应采样 最小梯度 预积分光照 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]