Scalable Approach for Content-Based Image Retrieval in Peer-to-Peer Networks

Abstract

Peer-to-peer networking offers a modular solution for sharing visual data across the network. it is a demand issue to perform CBIR in peer-to-peer networks. While most of the existing methods focus on LSH visual features and have limitations of flexibility. In proposed system we use an innovative approach that is a p2p network with CBIR by employing the BOVW model. As images are scattered across the p2p network, which makes a static codebook less effective for retrieval tasks. here we propose a codebook updating method by sharpen the mutual information between the follow -up codebook, and the workload balance among peers that manage different code words. In order to further improve retrieval performance and reduce network cost.