An incremental-learning neural network for the classification of remote-sensing images

A novel classifier for the analysis of remote-sensing images is proposed. S uch a classifier is based on Radial Basis Function (RBF) neural networks an d relies on an incremental-learning technique. This technique allows the pe riodical acquisition of new information whenever a new training set becomes available, while preserving the knowledge learnt by the network on previou s training sets. In addition, in each retraining phase, the network archite cture is automatically updated so that new classes may be considered. These characteristics make the proposed neural classifier a promising tool for s everal remote-sensing applications. (C) 1999 Elsevier Science B.V. All righ ts reserved.