Multimedia Content−based Information Retrieval

In this article we show an architecture which processes multimedial material, subdividing it on the basis of its own semantic contents. We propose an innovatory approach, whose methodology introduces an abstraction level, which has the task to study the relationships among the morphological attributes in a systematic way, before estimating the content. This analysis tries to unify the descriptors information and to gather them into structures that we call over−regions, which represent particular configurations of the objects to be recognized. The purpose is to facilitate the tasks reserved to the higher abstraction levels, down to the elements on which they must work won’t be low level attributes, but particular configurations in which those attributes place themselves. The format given to this analysis will have a case−based reasoning nature.