Web Services Based Qualitative Spatial Relation Analysis Component And Application

The spatial relation is a very important topic of the realm of spatial reasoning, Geographic Information System (GIS) and Computer Vision etc, and have already caused the strong interests of related realm researchers. Spatial relation is very important to query on a spatial database and conclude new information in the spatial reasoning. The topological relation of the spatial targets is the most basic spatial relation, and constituted the basic aspect of the spatial reasoning, some researchers even point out, the relation of the spatial targets and targets themselves have the equal importance, the research related with the topological relation occupy an important position in the spatial reasoning. In recent years, the development of the spatial relation's formal theories is shown in the fields of Artificial Intelligence, Geographic Information System (GIS), Computer Vision etc, and some research results already succeed applied in the practical applied realm.The work of this paper is that we advance a universal method of spatial relation analysis, and design and develop a WEB module -- the spatial relation analysis module base on logical reasoning uses this method, which is the complementarity of the national 863 high-tech project"Web Intelligence Applications Development and Runtime Platform"developed by the lab of knowledge engineering of Jilin University.First, this paper work for the formal show of the spatial relation model.The space relation is usually described by mathematics or the logic methods, and the purpose that given the formal description of the spatial relation model is to give the clear semantic description of the spatial relation, provide a formal tool for the spatial analysis and the spatial query. The development of the formal model of the spatial relation is a very important topic of the domain of the spatial reasoning, Geographic Information System (GIS) and Computer Vision, we choose three kinds of representative model of the spatial relation to analyze: The RCC formal model, the Broad Boundary model and the Direction model.Through the analysis and research of these three spatial relation models ,we found that, these three models have their own advantages at the spatial formal show aspects, but there is no universal method which can be used by every model, it means, the spatial relation analysis method of each model is independent. Therefore, we consider to design a new kind of universal spatial relation analysis method which can be used by every model in this paper. We know that the basic relationship of two spatial areas is the Overlap relationship (there is public parts in two spatial areas) and Contain relationship, and the other relationships of the two spatial areas can be get from the logic reasoning of this two relationships. According to this, we put forward the spatial relation analysis method that using logic reasoning. The main thought of that method is: First judge the two spatial relation values of two spatial areas for 0 or 1, and then get the other spatial relation value through logic reasoning .Then we give the logical formulas of each spatial relation model discussed in this paper.Secondly, we discuss the space the indetermination spatial model in the grid space.According to the characteristics of the indeterminate grid space, we discuss the modeling of the indeterminate grid space using fuzzy set method. For each indeterminate grid space area, each grid unit of the space area has three possible value: 0, 1 and between 0 and 1,each value represent that this grid unit belongs to the area, not belongs to the area, and possible belongs to de area. The characteristics of the indeterminate grid space has the same meaning with the definition of the degree function of the fuzzy set method, it means, we use the degree function represent the degree that each grid belongs to the spatial area.According to the definitions of indeterminate grid space area using method of fuzzy set, we give out the model of indeterminate grid space area based on the method of fuzzy set. We then give the two basic relationships shown as fuzzy set model, based on this , we give out the RCC model and the Taper Direction model's indeterminate logic formulas, it means, we use the degree functions of Overlap and Contain , through indeterminate logic reasoning , get the results of other relationships.Finally, we discuss the total design of the spatial relation analysis module and give the design and the development of method of spatial relation analysis.The spatial relation analysis component was developed based on the MVC structure, including the Spatial Model, the Deducing Control, and the User Interface. The description, reasoning and display of spatial relations are in different modules, so that the deducing process is no longer related with concrete models. We defined an interface criterion for spatial relations, so any model that matches the interface definition can be used in this component.The module of the spatial relation analysis based on the logic reasoning is a functional module, it has two main functions: The judgment of the basic spatial relation and logic reasoning based on the chosen model. The module of the judgment of basic spatial relation is a independent module, it has normative I/O interface and output the logic results of the two basic relations; And then the logic reasoning module use these results to reason and output the reasoning results to the user viewer.We discuss a new universal spatial relation analysis method - the spatial relation analysis method based on logic reasoning, and apply this method to the WEBGIS platform as a functional module. This method do not limit by the chosen spatial relation model, can judge the spatial relation for different model. The work of this paper has importance meaning for applying the Spatial Reasoning into applications of the Geography Information System. And there is still a lot of further research to complete in the future.