A Comprehensive Study of Periocular Biometrics on IRIS Recognition

Abstract

Among the best available biometrics like face, fingerprint, and iris, the iris is considered best due to being visible from a significant distance despite being an intrinsic part of the eye. Iris recognition for human authentication is already being implemented globally in various wide-range applications and is currently being researched for making it more secure and robust. Several covariates affect the overall performance of iris recognition. One such covariate is the periocular biometric which refers to the facial area in the immediate vicinity of the eyeball. In non-ideal conditions, where iris recognition fails, the periocular region comes as a savior. Periocular region can also act as a soft biometrics and adds to the accuracy and performance of the iris recognition system. This work first provides a timeline overview of the evolution of the biometrics for the sake of contextualization followed by a systematic and comprehensive review of periocular biometrics on iris recognition.

Keywords: Biometrics, Iris Recognition, Periocular Recognition, Covariate