Encoding and recognition of faces based on the human visual model and DCT

Abstract

For encoding and recognizing human faces in monochrome images, we propose a new method based on a combination of the discrete cosine transform (DCT), principal component analysis (PCA), and the characteristics of the Human Visual System. The novel aspect of the proposed non-Bayesian, approach is that, in the course of the recognition of face images, we also achieve image compression (in the form of encoding). With the help of examples, we demonstrate the superiority and advantages of the new method in comparison with the results found in the literature.