Extraction And Integration Of Spatial Attribute Data For Global Mapping Project

With the implementation of the “Belt and Road” and “Going Global” strategies,China urgently needs global spatial information to control global resource distribution and global socio-economic conditions.At present,the global mapping project provides the necessary foundation for the development of a global spatial data infrastructure.Automatic remote sensing image extraction technology or artificial vectorization is used to obtain high-precision global spatial data.However,the acquisition of attribute information has not had a satisfied solution.In recent years,the information technology has developed rapidly,and geographic information data has been displayed across the network in various forms,including commercial maps and Volunteered Geographic Information.Compared with traditional surveying and mapping data,they have the characteristics of rich information,wide coverage and fast update speeds.They have opened up a new time-efficient and low-cost way for the acquisition of global spatial information.Therefore,this paper takes the spatial vector data automatically extracted from remote sensing images and various types of geographic information data existing on the Internet as research objects,and explores the extraction and integration methods of spatial data attribute information.The main contents are as follows:(1)For the core terrain features such as roads and water systems,using existing Volunteered Geographic Information,different indicators are used to calculate the geometric features of polyline and polygon,and a method for integrating polyline attributes based on geometric matching and a method for polygon attributes based on geometric matching are proposed,achieving the integration of attributes via the matching relationship of features.(2)For the problem in(1),the limited attribute information of Volunteered Geographic Information leading to the low integrity of the integration result,the method of attribute integration using multi-source POI data is further studied.Firstly,a POI acquisition method and a method for extracting road names from POI address information are proposed.Then,a method for integrating the attributes of POI and polyline/polygon based on the proximal voting mechanism is proposed by calculating the proximity relationship between POI and the feature.For the inconsistence of polyline continuity and semantic information,an attribute integration method based on the generation of the principal curve is proposed.The POI generates the principal curve to simulate the geometry of the polyline.Then matching relationships are established through geometric matching to integrate attribute.(3)Aiming at the important facility that play an important role in national economic development and security such as airports and ports,this paper proposes to store information in the form of graph data.Relevant information is obtained from Volunteered Geographic Information and encyclopedia websites.Attribute information and relationships are extracted from them too.Then knowledge graph of important facilities is constructed.(4)A system is designed and developed to realize the methods proposed in this paper.It integrates the basic data visualization and attribute integration functions for experiments and data production.Then the algorithms are verified,and their results are analyzed.And an application to retrieve and visualize knowledge is designed and developed.