A neural network approach to geographic image analysis

Abstract

We have developed a method based on the precise pixel-to-pixel matching between two images. This is done by automatic generation of displacement vectors, carrying the information of differences between the two images. For generating a layer of vectors defining the information of displacement we use a neural network with self-learning architecture. The proposed algorithm perform successful mapping, which can be quantitatively measured as 90% correct recognition as demonstrated by the results. © 2000 IEEE.