A new method for face recognition based on PCA optimize strategy

This paper is focused on the problem of selecting optimum discrimination eigenvectors of PCA and improving the recognition accuracy. A new method for face recognition based on PCA optimize strategy is presented, in which the PSO algorithm is embedded, which select the recognition accuracy as the fitness value of particle swarm, to find out the optimum discrimination eigenvectors of PCA and obtain the optimal recognition accuracy simultaneously. We validate the effectiveness of this method with the ORL database and the Yale database. The experimental results indicate that the method can obtain the optimum discrimination eigenvector of PCA and a major improvement on recognition accuracy compared with the eigenvector selection approach based on the energy accumulative contribution rate.