Whenever a feature extracted from an image has a unimodal distribution, information about its covariance matrix can be exploited for content-based r etrieval using as dissimilarity measure the Bhattacharyya distance. To r educe the amount of computations and the size of logical database entry, we approximate the Bhattacharyya distance taking into account that most of the energy in the feature space is often restricted to a low dimensional subspace. The theory was tested for a database of 1188

more »

... xtures derived from VisTex with the local texture b eing represented by a 15-dimensional MRSAR feature v e ctor. The retrieval performance improved signi cantly relative to the traditional, Mahalanobis distance based approach in spite of using only one or two dimensions in the approximation.