Improved Face Recognition Algorithm Using Eigen Faces

Face recognition is one of the major challenge and is the most popular research areas in the computer vision. The research is focused to develop the computational model of face recognition that is fast, simple and accurate in different environments. In this study, we develop an improved recognition strategy based computational model to identify the face of an unknown person. The proposed computational model makes use of the well-known eigenfaces approach. A given set of face images are processed to compute eigenfaces. The eigenfaces are then projected into a number of unique features vectors. These features vectors form a database of vectors and are used to identify an unknown face. The feature vector of the unknown face is compared with vectors in the database using the improved recognition strategy described in this paper. The proposed approach is tested on a number of face images of a database which consists of 49 images of 9 persons. The results indicate the proposed recognition strategy through eigenfaces works well. Keywords— Feature vector, eigenfaces, eigenvalues, eigenvector, PCA