中文摘要：

在海量空间数据的组织与表达方面,与传统的空间数据模型与方法相比,地球剖分网格表现出了极大的优势。鉴此,本文提出了一种剖分网格下顾及用户兴趣的矢量地图多尺度表达方法,首先利用大量POI数据构建空间兴趣场模型,描述用户对地理信息的兴趣的空间分布;然后对空间要素进行分类分级,确定其表达的尺度范围;运用GeoSOT剖分理论划分不同层次的剖分面片,建立剖分面片与尺度的对应关系,根据用户对剖分面片所在区域的兴趣度大小,表达不同详细程度的空间要素,实现基于用户兴趣的空间数据多尺度表达。实验表明,该方法能够满足用户由概略到详细、由重要到次要的空间认知需求,同时能够获得较好的多尺度表达效果。

英文摘要：

Compared with the traditional spatial data model and method,global subdivision grid show a great advantage in the organization and expression of massive spatial data.In view of this,a method of vector map multi-scale representation considering user interest on subdivision gird is proposed.First,the spatial interest field is built using a large number POI data to describe the spatial distribution of the user interest in geographic information.Second,spatial factor is classified and graded,and its representation scale range can be determined.Finally,different levels of subdivision surfaces are divided based on GeoSOT subdivision theory,and the corresponding relation of subdivision level and scale is established.According to the user interest of subdivision surfaces,the spatial feature can be expressed in different degree of detail.It can realize multi-scale representation of spatial data based on user interest.The experimental results show that this method can not only satisfy general-to-detail and important-to-secondary space cognitive demands of users,but also achieve better multi-scale representation effect.