Towards Better Ocular Recognition for Secure Real-World Applications

In the last decade, biometric has become one of the most common uses for verification and authentication of someone's personal identity. Biometrics are defined as the measurement and analysis of unique physical or behavioral characteristics. Fingerprint is undoubtedly the most well-known biometric, but in the recent years ocular recognition has taken a huge jump forward. Ocular recognition (OR) is one of the most reliable personal identification methods in biometrics that can have flexible applications as opposed to the iris recognition. The basic breakdown of the eye is the iris, cornea, retina, pupil, sclera, and conjunctiva. The iris is where eye is actually colored; this is determined by the amount and type of pigment in one's iris. Iris recognition is an accurate form of ocular recognition. It works by using biometric systems to apply pattern recognition to a person's eye since no two people have the same patterns in their eye. Ocular recognition has been used as an identification method on both Apple and Android devices without much popularity. The reason for this is mainly related to the issues with the underlying algorithm used that affects the accuracy and security measures. Many different algorithms have been proposed and tested; some are more accurate than others. Our goal is to investigate different technologies currently used to help combine the best parts of both (Iris and Retina) in order to propose a better system for future implementation. With the proper design, ocular recognition could be a great tool for user authentication in emerging Internet of Things (IoT) and mobile markets.