| At present,with the expansion of GIS service areas from the Earth's surface to physical space,human economic and social space,information space and other virtual spaces,the types of associative relationship between entities that need to be described in the GIS field will be more abundant.The associative relationship between spatio-temporal characteristics has been unable to meet the new application requirements,and the associative relationship generated by the attributes of entity have received more and more attention.In the context of the new-generation GIS-pan-spatial information system,this paper discusses the associative relationship of multi-granularity spatiotemporal objects due to their attribute characteristics and discusses their concept theories,construction methods,and visualization methods.It provides a theoretical basis for multigranularity spatio-temporal objects modeling,and provides powerful information support for various association analysis,in-depth knowledge mining,and decision support applications based on the visualization of associative relationship.The main content of the paper is as follows:(1)This paper analyzes the status of the research on the relationship between entities,construction methods and visualization methods,and analyzes the problems existing in the research on the relationship between entity objects in GIS,and proposes the main research objectives and content of the paper.(2)This paper expounds the basic theory of multi-granularity spatio-temporal object attributive associative relationship,defines its concept,classification method,characteristics and specific composition content,studies the classification and basic calculation method of the relational strength calculation model,and proposes constraint rules for different relational functions.A method for describing the relationship between two attributes was proposed.(3)This paper designs data model and storage format for the attributive associative relationship of multi-granularity spatio-temporal objects,proposes the principle of constructing and designs the overall construction process.In the case without data source,this paper proposes a method of attributive associative relationship construction based on visual interaction selection,which includes a method for constructing a relational network supporting the reuse of relationship types,a method for selecting an associative object based on dynamic visual interface of objects,and a specific method for generating relationship network elements.In the case with data source,this paper proposes the data conversion batch construction method of attributive associative relationship,the construction steps are designed,and the specific methods and processes for each step are elaborated.(4)This paper analyzes the visual requirements of multi-granularity spatio-temporal objects' attributive associative relationship,constructs a visual expression system,designs the visualization process,analyzes the visual ambiguities in relational edge crossing nodes in the existing research of relational network visualization,and proposes a relational edge layout algorithm based on the cubic Bézier curve for node avoidance.The visual confusion in the direction visualization of the existing relation is analyzed.The relational edge direction display optimization based on the gradient rendering is proposed,and this method reduces the amount of graphics load and enhances the sense of orientation of the linear symbols.Three visual interaction methods for multigranularity spatio-temporal object attributive associative relationship are designed.(5)This paper designed and implemented a prototype system for multi-granularity spatiotemporal object-attributive relationship construction and visualization,designed the overall framework of the system,the attributive relationship model,and the content of the functional modules.Each module was tested,which verifies the rationality and feasibility of relevant research results. |
PDF Full Text Request