快速EZC-DCT地形压缩算法的并行优化
|
摘要
为能够在大规模地形实时渲染中提高渲染及数据压缩的速率,提出一种利用GPU并行优化的快速EZC-DCT地形压缩算法。采用二维快速DCT变换代替EZC-DCT算法中的DCT变换,在利用GPU对算法进行并行加速的基础之上,对算法的并行方案进行优化改进,更加有效地利用GPU强大的并行计算能力,分担CPU的负荷,快速完成相关计算。实验结果表明,该算法帧速率比原EZC-DCT方法提升约10个百分点,满足地形渲染的实时性要求。
To improve the speed of rendering and data compression in real-time rendering of large scale terrain,a fast EZC-DCT algorithm based on GPU parallel optimization was proposed.Two-dimensional fast DCT transform was used instead of DCT transform in EZC-DCT algorithm.On the basis of using GPU to speed up the algorithm,the parallel algorithm was optimized and improved.The powerful parallel computing ability of GPU was used,which not only shared the load of CPU,but also completed the correlation calculation quickly.Experimental results show that the proposed algorithm can increase the frame rate by 10% compared with original EZCDCT algorithm,and meet the requirements of real-time terrain rendering.
To improve the speed of rendering and data compression in real-time rendering of large scale terrain,a fast EZC-DCT algorithm based on GPU parallel optimization was proposed.Two-dimensional fast DCT transform was used instead of DCT transform in EZC-DCT algorithm.On the basis of using GPU to speed up the algorithm,the parallel algorithm was optimized and improved.The powerful parallel computing ability of GPU was used,which not only shared the load of CPU,but also completed the correlation calculation quickly.Experimental results show that the proposed algorithm can increase the frame rate by 10% compared with original EZCDCT algorithm,and meet the requirements of real-time terrain rendering.
引文
[1]ZHANG Junfeng,YAO Zhihong.LOD algorithm of terrain based on conservative screen error and isolated division of quad-tree[J].Journal of Southwest Jiaotong University,2013,48(4):666-671(in Chinese).[张俊峰,姚志宏.基于四叉树孤立分割和屏幕误差的地形LOD算法[J].西南交通大学学报,2013,48(4):666-671.]
[2]Li W,Chen Y,Wang Z,et al.An improved decimation of triangle meshes based on curvature[G].LNCS 8818:Rough Sets and Knowledge Technology.Springer International Publishing,2014:260-271.
[3]WEI Yingmei,GU Tianyang,YANG Bing,et al.Rapid continuous rendering technology of terrain based on wavelet[J].Computer Engineering,2007,33(19):218-220(in Chinese).[魏迎梅,谷天阳,杨冰,等.基于小波的地形连续快速绘制技术[J].计算机工程,2007,33(19):218-220.]
[4]Zhang JN,Shuai LI.LOD terrain rendering algorithm based on vertex texture[J].Journal of System Simulation,2008,20(7):1758-1764.
[5]Dong YF,Sun YK,Tang ZS.Interactive visualization of 3D lunar model with texture and labels,using Chang'E-1 data[J].Science China Physics,Mechanics&Astronomy,2013,56(10):2002-2008.
[6]DOU Mingliang,GAO Baolu,ZHANG Bing.Terrain compression drawing technology based on fast EZC-DCT[J].Application Research of Computers,2016,34(10):1-6(in Chinese).[窦明亮,高保禄,张冰.基于快速EZC-DCT的地形压缩绘制技术[J].计算机应用研究,2016,34(10):1-6.]
[7]Sun D,Li M.Crack repairing optimization of terrain rendering quadtree LOD algorithm[J].International Journal of Earth Sciences&Engineering,2014,7(6):2594-2600.
[8]Haweel RT,El-Kilani WS,Ramadan HH.Fast approximate DCT with GPU implementation for image compression☆[J].Journal of Visual Communication&Image Representation,2016,40(A):357-365.
[9]Coelho DFG,Cintra RJ,Kulasekera S,et al.Error-free computation of 8-point discrete cosine transform based on the Loeffler factorisation and algebraic integers[J].IET Signal Processing,2016,10(6):633-640.
[10]Zhai R,Lu K,Pan W,et al.GPU-based real-time terrain rendering:Design and implementation[J].Neurocomputing,2015,171(C):1-8.
[11]Mittal S,Vetter JS.A survey of CPU-GPU heterogeneous computing techniques[J].ACM Computing Surveys,2015,47(4):69.
[12]Chen X,Chang LW,Rodrigues CI,et al.Adaptive cache management for energy-efficient GPU computing[C]//Proceedings of the 47 th Annual IEEE/ACM International Symposium on Microarchitecture.IEEE Computer Society,2014:343-355.
[2]Li W,Chen Y,Wang Z,et al.An improved decimation of triangle meshes based on curvature[G].LNCS 8818:Rough Sets and Knowledge Technology.Springer International Publishing,2014:260-271.
[3]WEI Yingmei,GU Tianyang,YANG Bing,et al.Rapid continuous rendering technology of terrain based on wavelet[J].Computer Engineering,2007,33(19):218-220(in Chinese).[魏迎梅,谷天阳,杨冰,等.基于小波的地形连续快速绘制技术[J].计算机工程,2007,33(19):218-220.]
[4]Zhang JN,Shuai LI.LOD terrain rendering algorithm based on vertex texture[J].Journal of System Simulation,2008,20(7):1758-1764.
[5]Dong YF,Sun YK,Tang ZS.Interactive visualization of 3D lunar model with texture and labels,using Chang'E-1 data[J].Science China Physics,Mechanics&Astronomy,2013,56(10):2002-2008.
[6]DOU Mingliang,GAO Baolu,ZHANG Bing.Terrain compression drawing technology based on fast EZC-DCT[J].Application Research of Computers,2016,34(10):1-6(in Chinese).[窦明亮,高保禄,张冰.基于快速EZC-DCT的地形压缩绘制技术[J].计算机应用研究,2016,34(10):1-6.]
[7]Sun D,Li M.Crack repairing optimization of terrain rendering quadtree LOD algorithm[J].International Journal of Earth Sciences&Engineering,2014,7(6):2594-2600.
[8]Haweel RT,El-Kilani WS,Ramadan HH.Fast approximate DCT with GPU implementation for image compression☆[J].Journal of Visual Communication&Image Representation,2016,40(A):357-365.
[9]Coelho DFG,Cintra RJ,Kulasekera S,et al.Error-free computation of 8-point discrete cosine transform based on the Loeffler factorisation and algebraic integers[J].IET Signal Processing,2016,10(6):633-640.
[10]Zhai R,Lu K,Pan W,et al.GPU-based real-time terrain rendering:Design and implementation[J].Neurocomputing,2015,171(C):1-8.
[11]Mittal S,Vetter JS.A survey of CPU-GPU heterogeneous computing techniques[J].ACM Computing Surveys,2015,47(4):69.
[12]Chen X,Chang LW,Rodrigues CI,et al.Adaptive cache management for energy-efficient GPU computing[C]//Proceedings of the 47 th Annual IEEE/ACM International Symposium on Microarchitecture.IEEE Computer Society,2014:343-355.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |