METHODS FOR KNOWLEDGE DISCOVERY IN IMAGES

Abstract

In this paper we describe an automated method for generating images annotations, taking into account their visual features. The semantic rules map the combinations of visual characteristics (colour, texture, shape, position, etc.) with semantic concepts, capture the meaning and understanding of a domain by an expert, namely, which visual primitives are definitive for the semantic concepts of an image category. These rules are represented in Prolog and can be shared and modified depending on the updates in the respective domain.