The states of dilation in iris recognition (a preliminary study)

Abstract

The impact of pupil dilation on the matching accuracy of iris recognition algorithms has been demonstrated in the biometrics literature. However, the current literature does not model the various states of pupil behavior with respect to the underlying dynamics. Consequently, most existing work on this topic is empirical in nature. Our work uses concepts of transition processes and limiting distributions to describe the relationship between the state of the input iris image and a countable number of enrolled dilation states from an iris recognition standpoint. We also investigate a special case where a closed form expression is obtained that directly relates the various states to overall pupil behavior. Numerical evaluations demonstrate the feasibility of our proposed model where the results of our work can be directly used by iris recognition algorithms to account for pupil dilation artifacts.

Understanding the subject-specific effects of pupil dilation on iris recognition in the NIR spectrum

“…Our model can serve as an analytical description for describing pupillary response, which can be directly applied to explore the impact of dilation on iris recognition [20]. Furthermore, our process can be used to develop robust iris recognition algorithms that mitigate the effects of pupil dilation.…”

“…Our work is an advancement of the prior art because we provide an analytical basis for describing the effects of pupil motion that are also present in the NIR spectrum, which can be expanded to explore other dilation defects such as aging [17]. Also, our model can be applied to understanding the impact of dilation on iris recognition from a state perspective [20]. The rest of this work is presented as follows: Section 2 provides the theoretical foundation; Section 3 provides the experimental results and comparison; and Section 4 describes our conclusions and future directions.…”