计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (33): 167-172.
• 图形、图像、模式识别 • 上一篇 下一篇
王 军
出版日期:
发布日期:
WANG Jun
Online:
Published:
摘要: 基于图形处理器(GPU)上的多目标缓存提出了一种新的地质模型快速体素化算法,可以在GPU上绘制模型的过程中在像素着色器中利用多目标缓存上的逻辑或操作完成模型表面体素化处理,在后处理的屏幕绘制过程中对模型内部进行体素化填充,极大地提高了计算速度,使得大规模地质模型可以在常见GPU平台上得以实时体素化,可以较好地处理模型内部的空腔等问题。针对算法在模型相近表面或微小凹凸处会产生错误的情况,设计了一种基于逻辑异或操作的实体体素化改进算法,使其结果更为快速准确。
关键词: 地质模型, 体图形学, 体素化, 图形处理器, 多目标缓存
Abstract: This paper presents a fast method for voxelization of geological models based on the multiple render target buffers on the graphics processing units. This algorithm performs surface voxelization on the fragment shader using the logical OR operation, and fills the inside voxels in a following full-screen pass. It can correctly handle the holes inside the input models, and makes it possible to get real-time performance for large scale models on common graphics hardware. To alleviate the artifacts caused by adjacent surfaces or small concaves and convexes on the model surfaces, this paper also develops an improvement method using logical exclusive OR operation, which can generate more accurate results more efficiently.
Key words: geological model, volume graphics, voxelization, Graphics Processing Unit(GPU), Multiple Render Target(MRT)
王 军. 图形处理器上地质模型的快速体素化[J]. 计算机工程与应用, 2012, 48(33): 167-172.
WANG Jun. Fast voxelization of geological models on GPUs[J]. Computer Engineering and Applications, 2012, 48(33): 167-172.
0 / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://cea.ceaj.org/CN/
http://cea.ceaj.org/CN/Y2012/V48/I33/167