面向体数据的虚拟八叉树模型研究  

Study of virtual octree model for volume data

作　　者：吕广宪[1] 潘懋[1] 王占刚[1] 丛威青[1] 

机构地区：[1]北京大学地球与空间科学学院,北京100871

出　　处：《计算机应用》2006年第12期2856-2859,共4页journal of Computer Applications

摘　　要：针对常规八叉树和线性八叉树的不足,提出了一种基于多级Z-Order曲线、无指针/无位置码的虚拟八叉树模型。在时间方面,由于采用了规则划分的节点数据块及其简单高效的索引结构,新模型具有很高的内存访问效率;在空间方面,由于节点中无指针和位置码信息,而且采用了新的基于规则节点块的合并和压缩原则,新模型具有很好的存储效率。测试结果表明,虚拟八叉树模型同时具有指针八叉树在时间效率、线性八叉树在空间效率两方面的优势,是一种高效的三维体数据组织模型,在体图形学相关的领域中具有重要的研究意义和应用价值。Pointer-based octree and linear octree have their limitations either on time efficiency or storage efficiency. A pointerless/non-location-code virtual octree model based on multi-z-order-curves was proposed. In terms of time, the virtual octree provides fast access to the octree nodes by regularly partitioned node blocks and their index. In terms of storage, the virtual octree provides good compression by pointerless/non-location-code nodes and the node combination method based on regularly partitioned blocks. Test results show that the virtual octree has access efficiency as fast as pointer-based octree, and storage compression as good as linear octree. It is a good volume data model, and has wide application in volume graphics related areas.

关 键 词：空间信息景统 体图形学 体数据 八叉树 实体建模 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]