Resemblance of Biological Particle Swarm Optimization and Particle Swarm Optimization for CBFR by using NN

Abstract In this paper, we propose Biogeography Particle Swarm Optimization and Partical Swarm Optimization for CBFR which is based on Neural Network (NN) i.e. BPSONN approach for content based face retrieval. This learning based on biogeography of swarm migration is used for classical PSO to avoid trapping of local optima. BPSO is used to initialize the weights and threshold of neural network. CBFR measure function is based on color and texture features in which facial features is to be detected. Color histogram is used to find out color feature of an image and Filters are used to extract the texture features. Proposed method is tested on FEI dataset, to show illustrative and effective performance