In January 2000, Noah Kalina began taking a photograph of himself every day – and has continued for more than 20 years. A time-lapse video of his images can be seen online: EveryDay. These images, and images from similar projects, provide us with interesting opportunities to explore the effects of time lapse on iris recognition employed on images that were not originally intended for iris recognition. NIST obtained a license from Kalina to use a subset (7 half years from 2009-2015) of original, high resolution, digital images in biometric studies. This paper is our first published analysis of those images. Our license does not permit redistribution of the high resolution images. Our analysis demonstrates that over a period of at least 6 years, even though the majority of the images in this dataset did not provide solid matches, there are instances where mated iris image pairs from visible light images that were not obtained for the purpose of iris recognition match with match scores corresponding to a false match rate of 0:1% at a true accept rate of 1% . Though such performance is not useful in applications such as access control, there are cases where it may be useful. We note that the algorithms used in this analysis were not designed to work with the iris images that we extracted from the EveryDay project.
Iris Recognition on Noah Kalina's Everyday, Technical Note (NIST TN), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.TN.2154, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=931712
(Accessed June 17, 2021)