一种基于外存的海量地表离散点的交互编辑算法

An Out-of-Core Algorithm for Interactive Editing of Massive Discrete Points on Surface

提出了一种基于外存的索引数据结构,实现了对海量地表离散点的高效组织及实时绘制与交互.在分析现有对海量离散点数据或点云数据处理方法的基础上,采用规则四叉树与类区间树相结合的嵌套复合结构对数据进行组织,提高了数据的查询检索效率,从而实现了在不影响场景绘制显示的前提下,对数据进行的动态删除、添加和修改,更好地支持了数据的交互操作.对该索引组织结构进行了内外存相统一的设计与实现,并验证了该方法的正确性及有效性.

This paper presents an out-of-core index data structure to achieve the highly efficient organization of the massive discrete points on terrain surface and the real-time rendering and interaction. According to the existing methods of processing massive discrete points or large point clouds, nested composite structure using rule quad-tree and interval-like tree to combine organization is proposed for data organization which greatly improve the efficiency of data retrieval queries, and also implements the real-time dynamic operations like deletion, add and modification of data under the premise of not affeeting scene display in order to better support the rendering and interaction of data. Finally, the index structure of the unity of internal and external memory is design and achieved, and the correctness and validity of the method are verified.