An Implementation of Local Feature-based Iris Recognition

A recent emphasis on security has resulted in increased research attention being offered to the field of individual identification based on “biometrics”. A biometric feature is an inherent physical or behavioural trait that is unique among individuals. Easily acquired biometric features include fingerprints, gait, facial features, and voice. In addition to these, the human iris can also be considered a valid biometric feature for personal identification. The iris is the coloured ring on the human eye between the pupil and the white sclera. Each human iris has a unique “texture” of subtle features that varies greatly from person to person. Also, as noted by Daugman [4], iris features remain constant over an individual's lifetime and are not subject to changes produced by the effects of aging as other biometric features may be. For these reasons, the human iris is an ideal feature for highly accurate and efficient identification systems. Ma et al. [1] divide iris recognition techniques into four categories: phase-based methods [4], zero-crossing representation methods, texture analysis-based methods [2], and intensity variation analysis-based methods [1]. Due to the subtleness and great irregularity of iris features, the best method of iris recognition is still the subject of some debate (although both [1] and [4] claim very impressive results). For this project, I intend to conduct a brief survey of existing iris recognition techniques, followed by an implementation of Ma's intensity variation analysis-based approach for iris recognition [1].