Are iris crypts useful in identity recognition?

Abstract

We conducted an experiment in which participants were asked to annotate the crypts they see in iris images. The results were used to assess the utility of crypts in identity recognition for forensic applications. Although the inter-participant annotation consistency may be limited by crypts' noticeability and the range of genuine similarity scores obtained by the adopted matcher is correlated to the number of crypt pixels in the given image, the intra-iris crypt perception is sufficiently consistent for nearly 95% of query images to be correctly identified. The performance is expected to be better for the professional examiners with comprehensive training.

Iris crypts: Multi-scale detection and shape-based matching

“…By far DB1 is the only dataset in crypt research that has reliable human annotation data available. We avoided using 'challenging crypts' for training because [15] pointed out that human annotations for 'easy crypts' are much more consistent. The proposed system is evaluated on DB2 and DB3.…”

“…The parameters are selected through parameter sweeping and performance evaluation by comparing the detected crypts to human annotations. The images we used in parameter sweeping (DB1 in section 4.1) are part of the dataset used in [15] that have relatively strong crypts. From all the human annotations we selected the crypt pixels that are agreed more than half of the human examiners as the baseline to calculate eer 1 and eer 2 .…”

“…In [15], the authors use the overlapping percentage η to evaluate the similarity between feature images I k and I l :…”

“…DB1 and DB2 are from the human annotation experiment [15]. DB1 is the training dataset we used to select parameters for feature detection and matching.…”

“…It has been demonstrated in [15] that human examiners are able to consistently perceive crypts on iris images. The results were presented to support a human-in-the-loop framework for subject identification that may extend the applications of iris biometrics to the forensics field.…”