Reduction of semantic gap using relevance feedback technique in image retrieval system

摘要

This paper proposes a novel content based image retrieval system incorporating the relevance feedback technique. In order to improve the retrieval accuracy of content based image retrieval systems, research focus has been shifted in reducing the semantic gap between visual features and the human semantics. The five major techniques available to narrow down the semantic gap are: (a) Object ontology (b) machine learning (c) relevance feedback (d) semantic template (e) web image retrieval. This paper focuses on the relevance feedback technique by which semantic gap can be reduced in order to improve the retrieval efficiency of the system. The major challenges facing the existing relevance feedback technique is the number of iterations and the execution time. The proposed algorithm provides a better solution to overcome both these challenges. The efficiency of the system can be calculated based on precision and recall.