A neuro-fuzzy classifier for land cover classification

Abstract

In this paper, we present a neuro-fuzzy classifier derived from the generic model of a 3-layer fuzzy perceptron and implement a classification software system for land cover classification. Comparisons with the proposed and maximum-likelihood classifiers are also presented, We use the image of Daeduk Science Complex Town which is obtained by AMS (airborne multispectral scanner). The results show that the mixed composition areas such as "bare soil", "dried grass" and "coniferous tree" are classified more accurately in the proposed method. This system can be used to classify the mixed composition area like the natural environment of the Korean peninsula. This classifier is superior in suppression of the classification errors for mixtures of land cover signatures.