Thermal and visible sensor application: physiological thermal moment invariant analysis for infrared-based face identification

Abstract

Up to date methods and approaches in face identification field heavily depend on facial characteristics; such as location of eyes, length of nose and mouth. Ambient lightings have much influenced the visibility of these facial characteristics; where the visibility varies significantly with inconsistent external light source. In this paper, we present an extended framework for face identification based on thermal information extracted from facial images acquired from a Raytheon Palm-IR-Pro and Raytheon L-3 Thermal-Eye 2000AS sensored lense for themal images and Panasonic WV-CP234 for visible images. The inspiration initiating to this research is to engage in extracting significant facial characteristics from the acquired bio-thermal distribution information within a face, which differs from current facial characteristics that are visible over the skin. Encouraging results are produced which demonstrates the high capability of Hu's classical moment invariants as a feature in thermal based face identification and introducing new ways for classical methods to be further utilized in theoretical and empirical research area.