摘要
体绘制技术生成的图像中丢失了深度信息,已有的增强深度感知方法通常只针对某些结构区域,牺牲其它结构信息的同时又直接修改体绘制算法。面向光线投射体绘制算法,该文提出一种增强深度感知的方法,不直接修改光线投射算法。投影均匀网格到体数据表面,网格跟随表面变形后经光线投射绘制在结果图像中,用户根据变形网格能够感知图像中的深度信息。同时,为突显变形网格所反映的深度信息,对投影后的网格线进行深度相关的着色,并添加投影辅助线以连接不同深度表面上的投影网格。算法在统一计算设备架构下并行执行后,不仅能够实时生成图像支持用户的交互控制,且图像中增强深度感知的效果明显,特别是当体数据包含多个分离或者交叉物体时。
The depth information in volume data is lost in the image rendered by volume rendering technique. The existing methods of depth perception enhancement only enhance some structures in the volume data at the cost of other structures details, and they directly edit the volume rendering algorithm. For ray-casting algorithm, a method of depth perception enhancement is presented, and it does not directly edit the algorithm. Specifically, an inerratic grid is projected to the surface of volume data, and then the grid changing along surface is rendered in the final image. Users can apperceive the depth information of surface from the changed grid. Meanwhil, two methods are used to enhance the depth information of the grid projection lines, one is coloring the grid lines based on the depth, and the other one is adding accessorial lines to join the grid lines on two surfaces with different depths. When implemented using compute unified device architecture, the image is rendered in real-time under user interaction. The effect of depth perception enhancement in the final image is obvious especially when the volume data contains some disjunct or intersectant objects.
出处
《电子与信息学报》
EI
CSCD
北大核心
2015年第11期2548-2554,共7页
Journal of Electronics & Information Technology
基金
国家自然科学基金(61202129)资助课题~~
关键词
体绘制
深度感知
网格投影
并行执行
统一计算设备架构
Volume rendering
Depth perception
Grid-projection
Parallel implementation
Compute unified device architecture
