Eigenface approach based human face recognition

Abstract

This paper describe the principle of eigenface approach to detection and identification human face. In this approach, the features extracted from face images are individual, hidden, light insensitive, and effective to recognition. Wherever approach treats the face recognition as two dimensions recognition problems, taking advantage of the fact that faces are normally upright and thus may be described by a small set of two characteristics view. The face images for training and testing are selected carefully from Olivetti and Oracle Research Laboratory (ORL) face database in order to have minmum pose variation. The eigenface approach is proposed seems to be an adequate method to be used in face recognition due to its simplicity, speed, and learning capability.