Of Gis & Rs Integrated Data Structure And Indexing Mechanism

The traditional techniques to classify the remote sensing image are usually only based on the spectral features of the objects. Updating the GIS by the traditional method is to input the data by hand and needs practical surveying to get the geographical information. This approach consumes too much time and produces little effect,so GIS can not be updated in time. The integration of RS with GIS provides a new way to solve these problems. GIS can provide a large amount of prior knowledge about the landscape scenes,including the positions of the objects,the geometric properties,the spectral feature,the textural feature,the altitude,the information about the objects' distribution,etc. these kinds of information can help us to interpret the remote sensing image based on multiple rules,so that the interpretation can be automatic and the precision can be improved.Based on lots of advanced results of database,modern mathematics,RS,GIS,artificial intelligence and object oriented,etc. this thesis discussed the basic theory of GIS and spatial database,the method of spatial indexing method,the technique of transfer between raster data and vector data,the technique of GIS data integration platform,the spatial data structure of the integration of GIS and RS. Among these techniques and theories,this thesis emphasized on the following aspects:integrated spatial data structure based on the theory of topology;integrated spatial data structure based on the theory of object oriented;combined with knowledge mining representing a integrated spatial data structure based on semantic oriented which enhances the ability of system in knowledge expressing,intelligent inference and association.