GIS functions - Interpolation

Abstract

The main strength of GIS (Geographical Information Systems) is the common analysis of compound spatial and attributive data. These latter are collected by samples. The values of the collected type of data can be expanded to the sites where no samples are available using interpolation methods. The interpolation can be performed either in the phase of data production or when the data are used for spatial analysis. In the second case the interpolation is a GIS function. A brief review of interpolation methods is given with emphasis on the applicability of the particular method to one of the cases mentioned above. The possibilities of two new methods - the wavelet transformation and the artificial neural network (ANN) approach are also discussed. Especial attention is given to the possibilities of artificial neural networks in model building and interpolation. Connection between the data model and ANN approach is examined. In context of ANN the links between interpolation, classification and GIS functions are explained. As conclusions, several statements about GIS functions and modeling are given.