An automatic face recognition system in the near infrared spectrum

Abstract

Face recognition is a challenging visual classification task, especially when the lighting conditions can not be controlled. In this paper, we present an automatic face recognition system in the near infrared (IR) spectrum instead of the visible band. By making use of the near infrared band, it is possible for the system to work under very dark visual illumination conditions. A simple hardware enables efficient eye localization, thus the face can be easily detected based on the position of the eyes. This system exploits the feature extraction capabilities of the Discrete Cosine Transform (DCT) which can be calculated very fast. Support Vector Machines (SVMs) are used for classification. The effectiveness of our system is verified by experimental results. © Springer-Verlag Berlin Heidelberg 2005.