Research On Line Segment K Nearest Regions Based On Spatial Database

In recent years,it has become more and more important of spatial database query technology in the fields of intelligent identification system,geographic information system and so on.In spatial database,neighbor query is one of the important query types.However,the nearest neighbor query can not meet the actual needs,so the research direction of nearest neighbor query has changed from ideal environment to complex road network environment,and the research object has also changed from point object to line segment object.Because there is no solution to query irregular graphics in the existing nearest neighbor research,this paper focuses on the query method of k nearest regions of line segment based on European environment,obstacle environment and road network environment in spatial database.Firstly,in order to solve the problem that existing research results can not effectively deal with the k nearest regions of line segment in spatial database,a query method for k nearest regions of line segment based on Hilbert curve meshing is proposed.The HLNR query can shorten the subsequent query process and optimize the scanning process in the precomputation process.In the refining process,the accuracy of the query results is further improved.We further study the k nearest regions query method based on Voronoi diagram in obstacle environment.The query process is mainly divided into reduced data set,pruning process and refining process.This method optimizes query rate and accuracy in pruning and refining process according to Voronoi graph characteristics.Finally,a query method for the k nearest regions of line segment in road network environment is proposed,and the method is applied to update data set.Because the dynamic change of data sets may affect the query results,we study the method of line k nearest region query in the case of adding and deleting regions.The main processes of these two query algorithms are region set reduction and refinement.According to the decision rules,redundant queries can be avoided,and query efficiency and query accuracy can be optimized.