Pii: S0031-3203(01)00161-3

In image search based on chromatic similarity, the e.ectiveness of retrieval can be improved by taking into account the spatial arrangement of colors. This can serve both to distinguish images with the same colors in di.erent arrangement, and to capture the similarity between images with di.erent colors but similar arrangements. We propose a model of representation and comparison which attains this goal by partitioning the image in separate entities and by associating them with individual chromatic attributes and with mutual spatial relationships. The e.ectiveness of the proposed model is assessed in a user-based evaluation. Experimental results show the capability of the model to join and balance chromatic and spatial similarity, thus improving the e.ectiveness of retrieval with respect to representations based on a global histogram. ? 2002 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.