In this paper we describe the development of an image retrieval system that is able to browse, cluster and classify large digital image databases. This work was motivated by the projects of the Visualization Centre of the Eötvös Loránd University, where such problems are to be solved. The system's functions are based on a Gaussian mixture model (GMM) representation of the images. Image matching is done by the distance measure of the representations, based on the approximation of the Kullback-Leibler divergence of the GMMs. The GMMs are estimated with an improved expectation maximization (EM) algorithm that avoids convergence to the boundary of the parameter space. These form the basis of the clustering, where a variant of a genetic algorithm is used. The suggested algorithm is able to work with a large number of images or objects, the grid technology is a useful tool for generating several runs simultaneously.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.