A face recognition method based on a combination of integrated neural network and KICA algorithm

In order to raise the efficiency of face recognition, a method based on back-propagation (BP) neural network and probabilistic neural network (PNN) integration was introduced. The method uses the kernel independent component analysis (KICA) to extract facial features, puts eigenvectors into the BP neural networks and PNN to learn, and outputs the two classification and recognition results by relative voting method. This method effectively solves the interferences of illumination, facial expression, etc., and as a result improves the classification of the human face recognition ability.