Developed artificial neural network based on human face recognition

Face recognition has become one of the most important challenging problems in personal computer-human interaction, video observation, and biometric. Many algorithms have been developed in recent years. These algorithms are not sufficiently robust to address the complex images. Therefore, this paper proposes a soft computing algorithm based on face recognition. One of the most promising soft computing algorithms which is back-propagation artificial neural network (BP-ANN) has been proposed. The proposed BP-ANN has been developed to improve the performance of face recognition. The implementation of the developed BP-ANN has been achieved using the MATLAB environment. The developed BP-ANN requires supervised training to learn how to anticipate results from the desired data. The BP-ANN has been developed to recognize 10 persons. Ten images have been used for each person. Therefore, 100 images have been utilized to train the developed BP-ANN. In this research 50 images have been used for testing purpose. The results show that the developed BP-ANN has produced a success ratio of 82 %.