摘要
首先对地理空间的变化进行了研究 ,发现人们日常所处的空间 (如城市 )本质上是一个渐变空间 .在日常空间管理中 ,为空间实体建立动态空间索引 ,融合空间实体的层次关系和邻接关系 ,既维护了索引的完整性和一致性 ,又提高了系统的效率 .经过对现有空间索引结构的分析和比较 ,发现它们都缺少邻接关系的表达 ,只有将R树和Voronoi图结合起来 ,才能满足建立动态空间索引的要求 .最后给出了实现动态空间索引自我维护的插入、删除方法的实现步骤 .
After a research on the changes of Geo-Space, It is came to a conclusion that the space (city for example) people live in is a gradually-changed space in general. In the daily management of Geo-Space, much benefits can be obtained (such as maintaining the integrality of spatial index, getting a more efficient system ) if Geo-entities are dynamically indexed. After an analysis and comparison of the current spatial-data index structures, we found that they lack the presentation of proximity of Geo-data. So we proposed a new structure -R tree plus Voronoi diagram to index spatial data. Finally, we gave detailed methods on how to dynamically index spatial-data and realize self-maintennce (including insertion and deletion).
出处
《华南师范大学学报（自然科学版）》
CAS
2002年第2期38-44,共7页
Journal of South China Normal University(Natural Science Edition)
关键词
渐变空间
动态索引
R树
VORONOI图
GIS
gradually-changed space
dynamically spatial-data index
R trees
Voronoi diagram
