Principal component analysis on face recognition using artificial firefirefly swarm optimization algorithm

In the modern environment, the innovations emerging in information technology has driven us to focus on strengthening the security process. This led to the recent advancement in face recognition technology and special attention is given to the recognition process by applying a biometric system for personal identification. Face recognition is renowned as one of the efficacious applications of picture study, popularly applied for reliable biometric where security is the important quality attribute to be achieved. In this paper, a highly effective face recognition system has been proposed by incorporating genetic algorithms for better search strategy. The proposed model works in two-step processes: face feature extraction and face pattern matching. The Haralick features and features extracted from face databases using PCA are used for face recognition. The most eminent artificial firefirefly swarm optimization algorithm is employed for better searching and matching of facial features. From the simulation experiments performed on the faces warehoused in the OUR database, the result has shown that the model is highly efficient, the PCA method has achieved 80.6% of recognition rate and the AFSA has acquired 88.9% accuracy in correct recognition rate.