Face Recognition System Based On FFT Algorithm And Its Application

Face recognition is a research subject with a bright practical prospect and great challenges, which relates to some hot research fields such as digital image processing, pattern recognition and so on. Compare with other biometric technologies, such as finger print recognition and iris recognition, face recognition has characteristics of non-contact and friendly interface. So the users can agreeably accept this technology. Face recognition has grate significance of application in these fields such as public security tracing system, security validating system, files-management system, credit card testing system, video conference system and human machine communication system. The cruxes of face recognition technology are how to select proper on feature extraction algorithm and feature compare algorithm.This thesis focuses on the problem of face feature extraction and face feature compare in face recognition technology. First, face images are processed by a 2D-Fourier transform that has some effective properties such as a linear transform. The frequency feature of the transform of the face image is projected onto the two compressed orthogonal matrixes, which come from the SVD (Singular Value Decomposition) of the standard face image obtained by averaging all training samples and then the projecting coefficients are used as the algebraic feature of the face image. In the matching stage, the traditional minimum distance classification isimproved to recognize the unknown faces by using Euclidean distance as the similarity measurement.Finally, the standard face database includes face images with different expressions, different illumination conditions and different poses is selected to evaluate the face recognition. The experimental results have shown the effectiveness of the recognition rate, the recognition accuracy and its insensitivity to the facial expression, illumination and posture.