Key Technology Research Of Spatial Direction Relations

With the rapid development of cloud computing, mobile Internet, Internet ofthings, big data and other information technology, the global geographicinformation datas show explosive growth. Spatial database is the key technologyof geographic information system, which faces many problems of massivegeographic information data representation, storage, query, management and soon. In the geographic information data environment, the disadvantage oftraditional spatial database model is obvious. In recent years, many researchersstudied the data mining techonology which can mine the value of geographicinformation data, and achieved excellent results, but did not combined withtraditional methods.The traditional spatial data only describe the relationshiap of two spatialobjects. In a large geographic information dataes environment, the traditionalmethod has many problems about storage redundancy, retrieval slow and so on.In the paper, we proposed a algorithm whose name is density-based K-meansclustering algorithm, calculated the center node and the number of all types, andproposed a method which integrates the spatial orientation relationships andspatial clustering, which saves a lot of storage space and improves the speed ofquering.The direction relationship is an important brance of spatial relationship,which has become a hot research in recent years. Traditional Direction RelationalMo del did not support the whole spatial object (points, lines, surfaces) whichbased on the table look-up operation and is very inefficient and difficult tosupport the complex Direction Relation reasoning. Direction relation based onmatrix mod el provided a solution to solve these problems. That constituted thebasic theory for description, operator and reasoning of qualitative space relation,provided new id eas about space query and consistency check.