Skip to Main Content
Artificial neural networks have gained increasing popularity as an alternative to statistical methods for classification of remote sensed images. The superiority of neural networks is that if they are trained with representative training samples they show improvement over statistical methods in terms of overall accuracies. However if the distribution functions of the information classes are known, statistical classification algorithms work very well. To retain the advantages of both the classifiers, decision fusion is used to integrate the decisions of the individual classifiers. To overcome the difficulty of classification in high dimensional feature space feature fusion is done to reduce the dimensionality. AVIRIS images are used as test site and classification is initially achieved using maximum likelihood classifier followed by a set of neural network classifiers which include perceptron, hamming and hopfield networks. The decisions of these classifiers are fused in the decision fusion center implemented using second perceptron network. The results show that the scheme is effective in terms of increased classification accuracies.