Probabilistic combination of spatial context with visual and co-occurrence information for semantic image analysis

In this paper, a probabilistic approach to combining spatial context with visual and co-occurrence information for semantic image anal-ysis is presented. Overall, the examined image is segmented and subsequently an initial classification of the resulting image regions to semantic concepts is performed based solely on visual informa-tion. Then, a Genetic Algorithm (GA) is introduced for deciding on the optimal semantic image interpretation, realizing image analysis as a global optimization problem. The fundamental novelty of this work is that the GA incorporates in its evolutionary procedure a set of Bayesian Networks (BNs), which probabilistically learn the im-pact of the available spatial, visual and co-occurrence information on the final outcome for every possible pair of semantic concepts. Experimental results on two publicly available datasets demonstrate the efficiency of the proposed approach. Index Terms — Spatial context, bayesian network, genetic algo-rithm, semantic image analysis 1