Eye Biometric For Unconstrained Images In Visible Wavelength

Abstract

The eye biometric get rapidly increased attention specifically with visible wavelength clarification due to the increased accessibility of camera-based devices. Eye biometric is used to control access to restricted areas, it can be used in passenger control at airports. Major challenging objective in biometrics research is the development of recognition systems to work with unconstrained environments. one transitive research area aim to use visible wavelength (VW) light imagery to get data at importantly larger distances than normal and on moving subjects, which is a hard task because this real-world data is different from the NIR setup. The NIR wavelength is particularly dangerous because the eye does not respond to its natural mechanisms i.e, blinking, aversion and pupil compression. However, the use of visible light and unconstrained imaging setups can severely degrade the quality of the captured data, that increases the challenges in performing secure recognition. Eye biometric commonly done through iris recognition and sclera recognition. Iris features of the eye having unique characteristics for each human. Iris features are stable and permanent over human life, and environment effects cannot alter its shape. Sclera recognition has distinctive properties of its blood vessels. The blood vessels within the sclera have various distortion and shapes. Eye images captured under relaxed imaging conditions are taken from the UBIRIS.v2 database. In this paper we made a comparison of the results obtained from the implementation of existing algorithms for iris and sclera recognition systems independently, and then made a comparison of multimodal eye biometric systems.