Image Reconstruction for Face Recognition Based on 2DPCA and SVM

摘要

In contrast to the covariance matrix of principal component analysis (PCA), two-dimensional PCA (2DPCA) is easier to evaluate the covariance matrix accurately and has less time to determine the corresponding eigenvectors. Based on the high performance of 2DPCA in dimensional reduction and SVM in tackling small sample size, high dimension and its good generalization, a new method of face recognition based on 2DPCA+SVM is proposed in this paper. The computer simulation illustrates the effectivity of this method on the ORL database.