Labeling update of segmented images using conceptual graphs and Dempster-Shafer theory of evidence

The extraction of objects present in photographs is a major problem to tackle when considering photograph retrieval. Such objects (or people) have to be detected in a way to allow retrieval based on concepts and not on physical characteristics like colors or textures. To achieve a reasonable detection rate, in-context image analysis has to be used. We propose to use conceptual graphs (a knowledge representation formalism that allow fast processing) with Dempster-Shafer theory of evidence to update original labeling coming from a segmentation that labels image regions out of context. We explain the whole process and the results obtained on real home photos. The encouraging results obtained show the potential of our proposal