Data fusion of GIS and RS based on neural network
Yannan Sun, Wang Jiao-he · Dalian Ligong Daxue xuebao · 2005
Geographic information system(GIS) is a kernel technology of the earth observation.Remote sensing(RS) is an important external information source of GIS and a useful tool for renewing data.On the other hand,GIS can assist in analyzing RS data.Integration of RS and GIS is an important tool for collecting,managing and analyzing the spatial information.For this purpose now the scholars are studying the basal data structure for integrating the two types of data.A method of data fusion of GIS and RS is provided using the neural network with unchanging data memory structure based on users′ aim.A 5-layer network is constructed.Inputs of network are the characteristic values of RS and GIS.Then the distance between the input vector and center vectors and the value of the center vectors to determine weights of the network are adjusted.At last the output,the changed attribute value,is used to update database.The validity of this method is verified by supervising the land cover changing with Zhalong TM image and the vector map to update the attribute database.