Knowledge representation and interpretation of images guided by models: contribution of hybrid artificial intelligence

Introductory seminar by Isabelle Bloch, SCAI Chair of Research and Teaching in Artificial Intelligence (LIP6/ISIR)

The interpretation of images benefits from the modelling of knowledge about both the scene observed and the objects it contains and their relationships. We show in this context the contribution of hybrid artificial intelligence, combining different types of formalisms and methods. Thus spatial relations can be modelled by combining fuzzy sets and mathematical morphology. This knowledge can be introduced in structural models such as graphs, ontologies, logical knowledge bases, or neural networks, and the problem of image interpretation is then expressed as a problem of spatial reasoning. These approaches will be illustrated with examples in medical imaging, showing the interest of combining several approaches.