Context Driven Concept Based Image Retrieval

Several semantic image search schemes have been recently proposed to retrieveimages from the web. However, the query context is regularly ignored in thesetechniques and hence, many of the returned images are not adequately relevant.In this paper, we make use of context to further confine the outcome of thesemantic search engines. For this purpose, we propose a hybrid search enginewhich utilizes concept and context for retrieving precise results. In the proposedmodel, an ontology is exploited for annotating images and accomplishing searchprocess in the semantic level. Furthermore, the query of the user is modifiedwith the concepts available in the ontology. Next, we make use of search contextof the user and augment the query with the information extracted from the user’scontext to additionally eliminate irrelevant results. Experimental results showthat the combination of concept and context is effective in retrieving andpresenting the most relevant results to the user.