Model Based Approach for Content Based Image Retrievals Based on Fusion and Relevancy Methodology

This paper proposes a methodology for Content Based Image Retrievals (CBIR) using the concept of fusion and relevancy mechanism based on KL divergence associated with generalized gamma distribution to integrate the features corresponding to multiple modalities, feature level fusion technique is considered. The relevancy approach considered bridges the link to both high level and low level features. The target in the CBIR is to retrieve the images of relevancy based on the query and retrieving the most relevant images optimizing the time complexity. A generalized gamma distribution is considered in this paper to model the parameters of the query image and basing on the maximum likelihood estimation the generalized gamma distribution, the most relevant images are retrieved. The parameters of the generalized gamma distribution are updated using the EM algorithm. The developed model is tested on the brain images considered from brain web data of UCI database. The performance of the model is evaluated using precision and recall.