中文摘要：

针对海量机载LiDAR点云数据管理与可视化效率不高的问题,提出了一种四叉树和局部KD树相结合的混合空间索引结构以及内外存结合的数据调度模式。在全局,可以通过四叉树金字塔模型实现快速检索与调度;在局部,通过内存中构建的KD树实现高效的查询与显示。利用敦煌地区约10亿点的激光雷达数据进行了验证,达到30帧/s的显示效率,为大规模点云数据的可视化奠定了基础。

英文摘要：

This paper proposed a kind of hybrid index structure for organizing airborne LiDAR point cloud data to solve the problem of large data organization and visualization, which combines the global quadtree with local KD-tree. The global quadtree is used to index the upper level of point cloud data in global area. The KD-tree is used to index the data in a leaf of quadtree. This method not only can organize massive point cloud data, but also guarantee the indexing efficiency. To test our method, experiments using 1 billion points of Dunhuang area are conducted, which reach a 30 frame rate visualization speed, providing a primary step for large point cloud visualization.