大规模地形场景流式渐进传输  

Progressive streaming of massive terrain data

作　　者：施松新[1] 张引[1] 叶修梓[1] 张三元[1] 

机构地区：[1]浙江大学计算机科学与技术学院,浙江杭州310027

出　　处：《浙江大学学报（工学版）》2008年第11期1862-1867,共6页Journal of Zhejiang University：Engineering Science

基　　金：国家"863"高技术研究发展计划资助项目(2007AA04Z1A5;2007AA01Z311);教育部博士点基金资助项目(20060335114);浙江省科技计划资助项目(2007C21006)

摘　　要：结合基于小波变换的渐进几何压缩和流式传输技术,设计了全新的海量地形数据组织方式,提出了一种在网络环境下大规模三维地形数据的远程可视化解决方案.利用数字高程模型(DEM)数据的规整性,对大规模地形场景进行合适的网格划分,采用整数小波变换对分块地形网格进行数据去相关,然后利用改进的SPIHT算法对数据进行小波零树编码压缩,压缩数据被组织成多条渐进码流的形式.在交互渲染过程中充分利用帧间相关性,根据视点位置和网络带宽等相关参数对相应压缩码流进行多分辨率自适应控制传输,有效解决了海量数据远程传输的带宽瓶颈.A novel approach of wavelet-based progressive terrain data compression and streaming transmission was proposed, coupled with a new massive terrain data layout, which introduced a remote visualization solution for 3D terrain data over network. The large-area terrain mesh was partitioned into blocks utilizing the regularity of digital elevation mode （DEM） data, then a lifting scheme wavelet was used for the datasets decorrelation of each block. An improved SPIHT algorithm was used for zero-tree wavelet encoding, then the datasets were organized as progressive bitstreams after compression. The resolution of each visible terrain block could be calculated adaptively according to the viewpoint location and the network bandwidth during the interactive rendering. Utilizing the coherences between frames, the bottleneck of massive data remote transmission was solved effectively.

关 键 词：地形渲染 流式传输 小波压缩 零树编码 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TN919.81[自动化与计算机技术—计算机科学与技术]