Novel Indexing Method of Relations Between Salient Objects
R. Chbeir (Laboratoire Electronique Informatique et Image, Universite de Bourgogne, France), Y. Amghar (Laboratoire d’Ingeniere des Systemes d’Information, INSA de Lyon, France) and A. Flory (Laboratoire d’Ingeniere des Systemes d’Information, INSA de Lyon, France)
Copyright: © 2003
Since the last decade, images have been integrated into several application domains such as GIS, medicine, etc. This integration necessitates new managing methods particularly in image retrieval. Queries should be formulated using different types of features such as low-level features of images (histograms, color distribution, etc.), spatial and temporal relations between salient objects, semantic features, etc. In this chapter, we propose a novel method for identifying and indexing several types of relations between salient objects. Spatial relations are used here to show how our method can provide high expressive power to relations in comparison to the traditional methods.