Image classification by PCA and LDA based fuzzy neural networks

Since fuzzy neural networks (FNN) have been successfully applied to classification problems, this paper proposes a principal component analysis (PCA) and linear discriminant analysis (LDA) based FNN to achieve the image classification. In PCA, it can convert a set of observations into a set of linearly uncorrelated variables called principal components. In LDA, the weights are updated by seeking directions that are efficient for discrimination. In FNN, the parameter learning adopts the gradient descent method to reduce the cost function. Therefore, the proposed PCA-LDA-based FNN can efficiently classify highly confusable image patterns.