A fuzzy neural network model (FNN model) for classification using Landsat-TM image data

A fuzzy backpropagation neural network model is proposed. In this model a fuzzy weighted distance (FWD) is used to interpret the degree of possibility of the input pattern belongs to a certain category in the fuzzy domain. An extended backpropagation (EBP) algorithm is employed to make the conventional multilayer perceptron (MLP) model can be used in fuzzy classification procedure. At the end, the Landsat-TM image data are used to inspect the effect of proposed model.<>