基于球面退化四叉树格网的GlobeLand30数据表达及精度分析

Expression and Accuracy Analysis of Globe Land30 Data Based on DQG

Globe Land30为我国研制的世界上首套全球30 m地表覆盖数据集。该数据采用传统的通用横轴墨卡托投影(UTM)组织数据。为了解决地图投影给全球性问题带来的空间数据断裂及几何变形等问题,本文以球面退化四叉树格网(DQG)为基础,构建一种Globe Land30的全球连续表达模型;并选择相应的精度评价指标,对基于DQG模型表达的几何变形、数据误差的空间位置分布及定量化进行研究。结果表明:基于DQG模型既能保证Globe Land30数据表达的全球连续性,又能保持原有的数据精度,为Globe Land30在全球空间数据的可视化连续表达与深度分析提供了基础。

The world's first global land cover datasets at30-meters resolution(GlobeLand30)had been produced by National Geomatics Center of China(NGCC)in2014.This product is organized in tiles with the Universal Transverse Mercator Projection(UTM).In order to resolve the problem of data discontinuity and geometry deformation on the globe scale by map projection,a continuous and uniform expression model based on the spherical Degenerate Quadtree Grid(DQG)for the GlobeLand30data is presented in this paper.By choosing different representative sphere zones and corresponding error evaluation index,a series of contrast experiments are completed and analyzed for the error distribution and quantification of the GlobeLand30data in DQG model.The results show that the DQG model not only keep the expression of the GlobeLand30data continuous and uniform,but also keep the same accuracy as well,and will provide a foundation for the continuous visualization and depth analysis of the GlobeLand30appplications.