Abstract: To resolve the problem of the huge terrain data size and the inefficient data management, a management method of massive terrain data based on tile quadtree and Z-order and Hilbert filling curves is presented. The continuity of data is ensured by tile quadtree based on coarse-grained, which realizes terrain real-time rendering quickly. Non-pointer data index based on filling curves and data block’s coding is designed to realize the exact orientation between the coding of data block and the data block. In addition, terrain data prediction strategy is adopted to realize dynamical scheduling of terrain dataset in view field. The experimental results show that the algorithm in this paper can realize realistic real-time rendering of massive terrain dataset.

Key words: tile quadtree, terrain dataset, massive, filling curve, data index, Z-order, Hilbert

摘要： 针对大规模地形数据庞大数据管理效率低的问题，提出一种瓦片四叉树和Z型、Hilbert填充曲线相结合的海量地形数据管理方法。基于粗粒度的瓦片四叉树能够实现地形实时快速绘制并很好保证数据的连续性；依据Z型和Hilbert填充曲线和数据块编码设计的无指针数据索引实现数据块的快速准确定位。此外，数据预测策略能够实现可见数据的实时调度。经实验验证该方法能够实现海量地形数据真实实时绘制。

关键词: 瓦片四叉树, 地形数据, 海量, 填充曲线, 数据索引, Z型, Hilbert