基于GPU的流场数据体绘制技术研究
|
摘要
科学计算可视化是20世纪80年代后期提出并发展起来的一个新的研究领域。它的应用范围十分广泛,几乎涉及自然科学及工程技术的一切领域。计算流体力学就是科学计算可视化中重要的应用领域之一。随着计算机技术的不断进步和计算流体动力学(CFD)技术的不断发展,使得计算机的计算能力迅速提高,许多重要的图形、图像处理算法均可用硬件来实现,通过CFD数值计算获得的大量数据可以得到实时处理。
本文以基于GPU的流场数据体绘制技术研究为核心。首先,介绍了三维空间数据场可视化算法与可编程图形处理器,研究了典型的体绘制算法并比较它们的异同,探讨了用体绘制技术和可编程图形处理器对流场数据进行可视化的可行性和必要性。其次,重点研究了流场中标量场的可视化技术,在此基础上,实现了基于GPU的三维纹理映射体绘制算法和光线投射算法。最后,给出两种算法的绘制结果。
利用体绘制技术和可编程图形处理器对流场数据进行可视化,通过分析是可行的,取得了较好的绘制效果,提高了绘制速度。
Visualization in Science Computing is an important research field brought forward and developed from the later 80s of the last century, which has a broad application in most science and technology field such as Computational fluid dynamics (CFD) and so on. In the past two decades, significant progress has been made in the computer technical and CFD. A lot of important arithmetic for producing graphics and pictures from data, including the data from CFD, could be realized by hardware.
The Volume Rending for flow field data by GPU is the main task of the study. The 3D spatial data sets visualization arithmetic and GPU was introduced firstly. The typical Volume Rending arithmetic were described and compared. The feasibility and necessity to realize the visualization of flow field by Volume Rending and GPU were discussed. Secondly, the Visualization of scalar data sets was introduced as an emphasis and the volume rending arithmetic on the basis of GPU of 3D Texture Mapping and Ray-casting were carried out. The results by two arithmetic were given and analyzed.
The visualization of flow field by Volume Rending technique and GPU was feasible both in theory and practice. A better drawing within less time could be obtained in this study.
本文以基于GPU的流场数据体绘制技术研究为核心。首先,介绍了三维空间数据场可视化算法与可编程图形处理器,研究了典型的体绘制算法并比较它们的异同,探讨了用体绘制技术和可编程图形处理器对流场数据进行可视化的可行性和必要性。其次,重点研究了流场中标量场的可视化技术,在此基础上,实现了基于GPU的三维纹理映射体绘制算法和光线投射算法。最后,给出两种算法的绘制结果。
利用体绘制技术和可编程图形处理器对流场数据进行可视化,通过分析是可行的,取得了较好的绘制效果,提高了绘制速度。
Visualization in Science Computing is an important research field brought forward and developed from the later 80s of the last century, which has a broad application in most science and technology field such as Computational fluid dynamics (CFD) and so on. In the past two decades, significant progress has been made in the computer technical and CFD. A lot of important arithmetic for producing graphics and pictures from data, including the data from CFD, could be realized by hardware.
The Volume Rending for flow field data by GPU is the main task of the study. The 3D spatial data sets visualization arithmetic and GPU was introduced firstly. The typical Volume Rending arithmetic were described and compared. The feasibility and necessity to realize the visualization of flow field by Volume Rending and GPU were discussed. Secondly, the Visualization of scalar data sets was introduced as an emphasis and the volume rending arithmetic on the basis of GPU of 3D Texture Mapping and Ray-casting were carried out. The results by two arithmetic were given and analyzed.
The visualization of flow field by Volume Rending technique and GPU was feasible both in theory and practice. A better drawing within less time could be obtained in this study.
引文
[1]McCormick B H,DeFanti T A,Brown M D.Visualization in Scientific Computing.Computer Graphics[J],1987,21(6).
[2]唐泽圣.三维数据场可视化[M].北京:清华大学出版社,1999.
[3]许元飞.三维数据场可视化中体绘制技术的研究[D].西安科技大学,2005.
[4]缪琳.三维数据场的信息处理与体绘制技术[D].国防科学技术大学,2004.
[5]石教英等.科学计算可视化算法与系统[M].科学出版社,1996.
[6]W.L.Cai,T.Z.Chen and J.Y.Shi.Rendering of Surface and volume details in volume data[J].Computer Graphics Forum,1995,14(3):4211-4230.
[7]H.E.Cline,W.E.Lorensen.Two algorithms for three-dimensional reconstruction of tomograms[J],medical physics,1998,15(3):320-327.
[8]Levoy M.Display of Surface from Volume Data.IEEE Computer Graphics and Application,1988,8(3):29-37.
[9]任静.基于D3D的体绘制技术研究[D].西安建筑科技大学,2007.
[10]洪伟,刘亚妮,李骑,丁莲珍.Cg教程-可编程实时图形权威指南[M].人民邮电出版社,2004.
[11]董朝.基于可编程图形硬件的若干技术研究[D].浙江大学硕士学位论文,2005.
[12]罗艳.基于可编程图形处理器的体绘制技术研究[D].西南交通大学,2007.
[13]Ecker M.Programmable Graphics Pipeline Architectures.http://xengine.sourceforge.net,2003.
[14]Zenz D.Advances In graphics Architectures.Dell Corp,Technical white Paper.http://www.dell.com,2002.
[15]邓郑祥译.OpenGL编程指南[M].人民邮电出版社,2005.
[16]Mason Woo,JacMe NeideL Tom Dawis etc.The offical guide to learning OpehGL[M].北京:中国电力出版社,2001.
[17]王翱宇.非感实感渲染若干技术的研究[D].浙江大学硕士学位论文,2005.
[18]Mark Willliam R,Glanville R steven,Akeley kurt.Cg:A system for Programming graphics hardware in a C-like language.ACM Transaction on Graphics,2003.
[19]Cg Toolkit User's Manual.A Developer's Guide to Programmable Graphics(Release 1.4).NVIDIA Corporation,2005.
[20]Lorensen,W E.Cline,H.E.Marching Cube:A High Resolution 3D Surface Construction Algorithm.Computer Graphics,1987,21(4).
[21]Doi A,koide A.An Efficient Method of Triangulating Equi Value Surfaces by using Tetrahedral cells.IEICE transactions,1991,E74(1).
[22]Cline H E,lorensen W E.Two algorithms for three-dimensional reconstruction of tomograms[J].Medical physics,1988,15(3).
[23]Gallagher R S,Nagtegal Jc,karlsson H,et al.An Efficient 3D visualization Technique for Finite Element Models and Other Coarse Volumes.Computer Graphics,1989,23(3).
[24]Shepard D.A Two Dimensional Interpolation Function for Irregularly Spaced Data.Proceedings of ACM 23rd National Conference,1968:517-524.
[25]Wilhelms J,Challinger J,Alper N,et al.Volume Rendering of Curvilinear Volumes.Computer graphics,1990,24(5):41-47
[26]Koyamada K,Nishio T.Volume Visualization of 3D Finite Element Method Results.IBM Joural Res.Develop,1991,35(1/2):12-25
[27]池峰.医学影像容积重建及可视化技术研究[D].西安电子科技大学,2004.
[28]陈寅秋,董金祥,王晓梅.一个新的基于3D纹理映射及Shear-Warp交换的快速体绘制方法[J].计算机工程,1998,24(8):14-19.
[29]储璟骏.基于GPU的直接体绘制技术[D].上海交通大学,2007.
[30]Nvidia.The Geforce 6 series GPU Architecture.Gpu Gems2,2005.
[31]Moccia D,Cg users Mannual.http://download.nvidia.com/developer/cg/cgl.4/1.4.1.UserManual.pdf,2005.
[32]吕珍.基于OpenGL的流场数据可视化技术[D].武汉理工大学,2007.
[33]田玲.基于体绘制的三维数据场可视技术研究[D].西南交通大学,2007.
[34]Macedonia Michaell.The GPU enters computing's maintream.IEEE Computer,2003,36(10):106-108.
[35]杨兵,李凤霞,战守义,胡敏勇.GPU在复杂场景的阴影绘制中的应用[J].计算机工程,2006,32(2):220-222.
[36]柳有权,刘学慧,吴恩华.基于GPU带有复杂边界的三维实时流体模拟[J].软件学报,2006,17(3):568-576.
[37]唐慧,周正东,鲍旭东,罗立民.基于GPU的三维医学图像混合可视化系统[J].数据采集与处理.2006,21(4):428-433.
[38]张延红,周必水.基于GPU加速的光线与三角面片求交[J].计算机时代.2006,5:29-31.
[39]张建勋,刘全利,陈庄.基于可编程GPU的快速体绘制技术[J].重庆大学学报,2005,28(7):67-70.
[40]吴恩华,柳有权.基于图形处理器(GPU)的通用计算[J].计算机辅助设计与图形学报,2004,16(5):601-612.
[2]唐泽圣.三维数据场可视化[M].北京:清华大学出版社,1999.
[3]许元飞.三维数据场可视化中体绘制技术的研究[D].西安科技大学,2005.
[4]缪琳.三维数据场的信息处理与体绘制技术[D].国防科学技术大学,2004.
[5]石教英等.科学计算可视化算法与系统[M].科学出版社,1996.
[6]W.L.Cai,T.Z.Chen and J.Y.Shi.Rendering of Surface and volume details in volume data[J].Computer Graphics Forum,1995,14(3):4211-4230.
[7]H.E.Cline,W.E.Lorensen.Two algorithms for three-dimensional reconstruction of tomograms[J],medical physics,1998,15(3):320-327.
[8]Levoy M.Display of Surface from Volume Data.IEEE Computer Graphics and Application,1988,8(3):29-37.
[9]任静.基于D3D的体绘制技术研究[D].西安建筑科技大学,2007.
[10]洪伟,刘亚妮,李骑,丁莲珍.Cg教程-可编程实时图形权威指南[M].人民邮电出版社,2004.
[11]董朝.基于可编程图形硬件的若干技术研究[D].浙江大学硕士学位论文,2005.
[12]罗艳.基于可编程图形处理器的体绘制技术研究[D].西南交通大学,2007.
[13]Ecker M.Programmable Graphics Pipeline Architectures.http://xengine.sourceforge.net,2003.
[14]Zenz D.Advances In graphics Architectures.Dell Corp,Technical white Paper.http://www.dell.com,2002.
[15]邓郑祥译.OpenGL编程指南[M].人民邮电出版社,2005.
[16]Mason Woo,JacMe NeideL Tom Dawis etc.The offical guide to learning OpehGL[M].北京:中国电力出版社,2001.
[17]王翱宇.非感实感渲染若干技术的研究[D].浙江大学硕士学位论文,2005.
[18]Mark Willliam R,Glanville R steven,Akeley kurt.Cg:A system for Programming graphics hardware in a C-like language.ACM Transaction on Graphics,2003.
[19]Cg Toolkit User's Manual.A Developer's Guide to Programmable Graphics(Release 1.4).NVIDIA Corporation,2005.
[20]Lorensen,W E.Cline,H.E.Marching Cube:A High Resolution 3D Surface Construction Algorithm.Computer Graphics,1987,21(4).
[21]Doi A,koide A.An Efficient Method of Triangulating Equi Value Surfaces by using Tetrahedral cells.IEICE transactions,1991,E74(1).
[22]Cline H E,lorensen W E.Two algorithms for three-dimensional reconstruction of tomograms[J].Medical physics,1988,15(3).
[23]Gallagher R S,Nagtegal Jc,karlsson H,et al.An Efficient 3D visualization Technique for Finite Element Models and Other Coarse Volumes.Computer Graphics,1989,23(3).
[24]Shepard D.A Two Dimensional Interpolation Function for Irregularly Spaced Data.Proceedings of ACM 23rd National Conference,1968:517-524.
[25]Wilhelms J,Challinger J,Alper N,et al.Volume Rendering of Curvilinear Volumes.Computer graphics,1990,24(5):41-47
[26]Koyamada K,Nishio T.Volume Visualization of 3D Finite Element Method Results.IBM Joural Res.Develop,1991,35(1/2):12-25
[27]池峰.医学影像容积重建及可视化技术研究[D].西安电子科技大学,2004.
[28]陈寅秋,董金祥,王晓梅.一个新的基于3D纹理映射及Shear-Warp交换的快速体绘制方法[J].计算机工程,1998,24(8):14-19.
[29]储璟骏.基于GPU的直接体绘制技术[D].上海交通大学,2007.
[30]Nvidia.The Geforce 6 series GPU Architecture.Gpu Gems2,2005.
[31]Moccia D,Cg users Mannual.http://download.nvidia.com/developer/cg/cgl.4/1.4.1.UserManual.pdf,2005.
[32]吕珍.基于OpenGL的流场数据可视化技术[D].武汉理工大学,2007.
[33]田玲.基于体绘制的三维数据场可视技术研究[D].西南交通大学,2007.
[34]Macedonia Michaell.The GPU enters computing's maintream.IEEE Computer,2003,36(10):106-108.
[35]杨兵,李凤霞,战守义,胡敏勇.GPU在复杂场景的阴影绘制中的应用[J].计算机工程,2006,32(2):220-222.
[36]柳有权,刘学慧,吴恩华.基于GPU带有复杂边界的三维实时流体模拟[J].软件学报,2006,17(3):568-576.
[37]唐慧,周正东,鲍旭东,罗立民.基于GPU的三维医学图像混合可视化系统[J].数据采集与处理.2006,21(4):428-433.
[38]张延红,周必水.基于GPU加速的光线与三角面片求交[J].计算机时代.2006,5:29-31.
[39]张建勋,刘全利,陈庄.基于可编程GPU的快速体绘制技术[J].重庆大学学报,2005,28(7):67-70.
[40]吴恩华,柳有权.基于图形处理器(GPU)的通用计算[J].计算机辅助设计与图形学报,2004,16(5):601-612.