Subtractive clustering for PCA image coding

Abstract

Principal component analysis (PCA), a well-known statistical processing technique, allows to research the correlation among the components of multi-dimensional data and to reduce redundancy by the projection of data over a proper orthonormal basis. In this paper, we employ PCA for image compression and adopt the neural network architecture in which the synaptic weights, served as the principal components, are trained through generalized Hebbian algorithm (GHA). In addition, we partition the training set into clusters using the subtractive clustering method obtain better retrieved image qualities.