Identical Entity Matching Method Of Multi-scale Land Use Data

Spatial data is the research object and foundation of geographic information science. Spatial data integration and update are important means to protect the investment of data collection. There are significant differences between data at different times and with different collecting purposes, which become a big inconvenience to application of data integration and update. In the process of updating, a lot of manual interventions are required because there are spatial and attributive inconsistencies between updating data and fundamental data. In the process of multi-scale spatial vector data integration, the data cannot form coherent dataset because of the model, scale and classification differences. Identical entity matching is the key technology to solve this problem.At present, there have been some proven methods and techniques to solve the problem of identical point matching in raster data and simple geometric identical entity matching. But the current study failed to give preferable matching result in the complex matching situation. Meanwhile, there are less research in matching index selection and process optimization. Against the problems in the current study, this paper starts with the practical issues of multi-scale land use related data matching, looks into the spatial entity, especially polygons, matching methods, design and optimize similarity index and matching process, to improve the efficiency of multi-scale land use data matching.The main contents and conclusions of this paper include:(1) Elaborate the problems faced by the spatial data updating and integration, as well as the supporting role of identical entity matching technology. Summarize the research progress, technical difficulties and trends in current domestic and international study. (2) Based on the data of study area, the difference of land use related data, its formation causes and effects to data analyzing are stated. The matching problems in land use change survey, land use data temporal analysis and data integration are analyzed. Against each business part, the difference of data type, data source and main matching basis are discussed. (3) Based on the analysis of the problem, the geometric and topological similarity index are summarized; According to the actual demand of multi-scale land use data matching, analysis, the polygon difference characteristics are found out, the method to measure similarity of overall and local shape is raised, and the spatial location similarity is measured by conversion index. (4) Based on the similarity index analysis, the rules of identical entity matching process are studied:Through the analysis of spatial object position uncertainty, put forward similarity measurement method considering uncertainty; iterative strategy and multi index judgment strategy in multi similarity index matching are discussed; propose the candidate objects retrieval method in matching process. (5) Realize the identical entity matching method and technical process through programming. Empirical research are carried out using land use data and agricultural land grade data, the result shows that the method proposed in this paper can overcome the problem in multi scale land use data and improve matching efficiency.