Efficient Hybrid Structure of Spatial Index for GIS Databases
Chen Tian-zi
Abstract
Chen Tian-zi
Abstract
According to the characteristics of spatial index of grids and R-tree,this paper proposes a hybrid structure of spatial index based on wide grids and clustering R-tree(HSIBGR).In HSIBGR,a rectangular region of geography is roughly partitioned to several sub-regions,which have same areas and shapes.For objects in every sub-region,we construct spatial index of R-tree.Nodes splitting of R-tree are implemented based on clustering techniques.The authors analyze performances of HSIBGR though theoretical methods and evaluate those by experimental results.The results show that HSIBGR can effectively reduce overheads of storages and improve efficiency of searching.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
According to the characteristics of spatial index of grids and R-tree,this paper proposes a hybrid structure of spatial index based on wide grids and clustering R-tree(HSIBGR).In HSIBGR,a rectangular region of geography is roughly partitioned to several sub-regions,which have same areas and shapes.For objects in every sub-region,we construct spatial index of R-tree.Nodes splitting of R-tree are implemented based on clustering techniques.The authors analyze performances of HSIBGR though theoretical methods and evaluate those by experimental results.The results show that HSIBGR can effectively reduce overheads of storages and improve efficiency of searching.
Key concepts: Spatial database, Computer science, R-tree, Index (typography), Data mining, Cluster analysis, Spatial analysis, Tree (set theory)