Kernel function is the function which computes dot product in feature spaces. Both the SVMs and kernelPCA are kernel-based learning methods. In this paper, the SVMs and kernel PCA are used to tackle the face recogni-tion problem. SVMs are classifiers which have demonstrated high generalization capabilities. Kernel PCA is a featureextraction technique which is proposed as a nonlinear extension of a PCA. We illustrate the potential of SVMs andkernel PCA on the Yale database and compare with a PCA based algorithm. The experiments indicate that SVMs andkernel PCA are superior to the PCA method.
Statistical learning theory(SLT) and support vector machine(SVM) are effective to solve problems of machine learning under the condition of finite samples. It is known that the performance of support vector machine is often better than that of some neural networks in pattern recognition,especially in high dimensional space, and they are well used in many domains for recognition. This paper at first introduces the basic theory of SLT and SVM,then points out the key problems of SVM and its research situation in recent years,and at last describes some applications of SVM in the field of pattern recognition.
