Face Recognition using Eigenfaces, PCA and Supprot Vector Machines

This paper is based on a combination of the principal component analysis (PCA), eigenface and support vector machines. Using N-fold method and with respect to the value of N, any person’s face images are divided into two sections. As a result, vectors of training features and test features are obtained. Classification precision and accuracy was examined with three different types of kernel and appropriate number of face features was considered and the best function for system identification rate. Then, face features were fed into the support vector machine (SVM) with one vs. all classification. At first, 2-Fold method was examined for images of training and test system. The results indicated that the rotation of the sets in identical classifications had no impact on the efficiency of radial basis function (RBF). It was observed that the precision increased in the 5-Fold method. Then, 10-Fold method was examined which indicated that the average recognition rate further increased when compared with 2-Fold and 10-Fold methods. The results revealed that as the rotation number increases, the precision and efficiency of the proposed method for face recognition increases.