Face Recognition Method Based On Principal Component Analysis, Artificial Neural Networks

Face Recognition is one of the key issues in pattern recognition and artificial intelligence fields. It involves image processing, psychology, physiology, recognition science and so on, which has a wide range of potential applications in the areas of identification of certificate, security control and criminal capture.This thesis including:(1) After analyzing the application of principal component analysis and prevailing face recognition technology, a multi-pose approximate frontal face recognition method based on principal component analysis (PCA) and back propagation (BP) neural network is proposed. By the analyses of different person and the same person with different expression and gesture, the number of principal component vector is determined properly under the attention to divisibility and workload. The features extracted by the selected PCA components are used in neural networks to finish the person recognition.(2) An improved back propagation (BP) algorithm was introduced to train the neural network for recognition. Comparing with the traditional BP algorithm, the improved algorithm can promote the learning rate and credibility of the algorithm. It adopts momentum method and learning rate adaptive adjustable strategy. The momentum method lower the sensitivity to details of error curved surface and efficiently restrains from falling into the local minimum. Adaptive adjustable strategy can effectively shorten the learning time and accelerate convergence rale. Finally, the features of the images were extracted by PCA were sent into the improved BP neural network for classification and recognition.