Research On Face Recognition Based On Wavelet Transform And Support Vector Machines

With the fast development of the technology on the face recognition, especially under the rapid spur of computer, computer-aided face recognition has become the latest one among all the face recognition technologies. As a non-rigid object, human faces have the characteristics of vast variations, being prone to be affected by many influential elements. This paper proposes the approach in integration of combining wavelet transform and support vector machines. Wavelet transform can fully demonstrate the characters of the target problem and it enjoys the outstanding capability of obtaining comparatively less traits and insensitive to the facial expressions in the face detection. Support vector machines (SVM) are specially devised to solve the small sample problem. Based-on the structural risk minimization principle in the statistical learning theory, SVM selects the optimal separate hyperplane as the separate function. The optimal separate hyperplane is the hyperplane that either correctly separates the sample set or gets the biggest margin between two classes. Thus the separate problem can be formulated as a quadratic optimization problem satisfied simple restriction. The quadratic problem has singular global maximum point. By introducing the kernel function, the nonlinear separate samples are projected into a high dimension space (so call "feature space"), thus separate problem is solved in the linear separate feature space. SVM exhibits some particular superiority for the classification of small samples and has been the preferred classifier at the international level. The thesis uses wavelet transform to extract face image feature, then classifies them based on SVM. Recognition results of the experiments on ORL Face Database demonstrate the effectiveness of the system.