Improving the precision of CBIR systems by feature selection using binary gravitational search algorithm

In this paper, feature selection using binary gravitational search algorithm is utilized to improve the precision of CBIR systems. Content-based image retrieval, CBIR, is one of the most challenging problems in the field of pattern recognition. The performance of a CBIR system is hardly depends on the features that are extracted from images. Thus, selecting most relevant features leads to higher accuracy by reducing the semantic gap between high level features and low level features. Gravitational search algorithm is one of the recent heuristic search algorithms that in this paper, its power is compared with genetic algorithm and binary particle swarm optimization in feature selection. The proposed method is examined in Corel database. Results confirm the efficiency of BGSA to increase the precision of CBIR systems.