球面栅格Voronoi图既具有自然邻近、动态稳定等优良性质，又符合计算机离散化特征，成为全球海量空间信息多尺度管理与分析最有潜力的解决方案之一。但是，由于空间栅格距离与欧氏距离的差异，传统球面栅格Voronoi图算法存在缺陷，其建模质量也具有一定的不确定性，极大地限制了其应用的范围与深度。为此，本项目针对这一学术前沿问题，在全球四元三角格网模型基础上，对球面多尺度栅格Voronoi图生成算法及可靠性评价模型进行探索，主要内容包括：基于QTM三角扫描的球面Voronoi图精确生成算法、格网Voronoi图多尺度计算与融合方法、球面格网Voronoi图误差评价与控制建模。研究成果有望为高精度球面格网Voronoi图的生成，提供高效精确的算法及可靠的质量评价模型，为大范围、不同尺度下球面格网Voronoi图的应用提供可靠保障。

The spherical raster Voronoi diagram, which has the property of natural nearby, dynamic stability and accords with computer discretization characteristics, has become one of the most potential solutions to manage and analyze multi-scale global spatial data. However, because of the difference between raster distance and Euclidean distance, there are some limitations and indeterminacy of traditional spherical raster Voronoi diagram generating algorithms and its modeling quality that have greatly limit the scope and depth of its application. Multi-scale raster spherical Voronoi diagram generating algorithms and reliability evaluation model will be researched in this project on the basis of the global Quaternary Triangular Mesh(QTM). The main contents include the generating algorithm based on QTM triangular-scans, multi-scale computing and fusion mechanism, error evaluation and control model of spherical grid Voronoi diagram. The research is expected to provide an efficient and accurate generating algorithm, a reliable quality evaluation model for spherical Voronoi diagram, which will provide a reliable guarantee for the multi-scale application of Voronoi diagram in large areas.