Abstract: With the arrival of the era of geographic information big data, geographic information and multiple industries are increasingly deeply integrated, resulting in a large number of heterogeneous multi-source data.Multi-source and rich spatial data provide basic support for economic and social development, while its heterogeneity also brings problems and challenges to data sharing and application.In view of the current situation and characteristics of multi-source, heterogeneous and inconsistent geospatial data, this paper puts forward the process and method of multi-source geospatial data fusion, which integrates database model fusion and database instance fusion, and analyzes the case data such as fundamental geographic information and national census geography, which provides a solution for multi-source spatial data sharing and application.

Key words: multi-source, geospatial data, database, data fusion