Effective Color Image Retrieval Based on the Gaussian Mixture Model

The main problem addressed in this paper is as follows: a system applying the proposed framework should retrieve all images whose color structure is similar to that of the given query image, independently on the applied lossy coding. We propose an approach based on the color histogram approximation using the Gaussian Mixture Model. The proposed method incorporates the information on the spatial distribution of the color image pixels utilizing the bilateral filtering scheme. The retrieval results were evaluated on large databases of natural color images and the usefulness of the proposed technique was compared with some commonly known retrieval methods operating on color histograms.