International. A report by the National Institute of Standards and Technology (NIST) was recently released with details on the evaluation of iris recognition software from eleven different entities. He indicated that this method of identification is being easier and faster, however it is true that its results are less accurate.
The Iris Exchange (IREX) III report is the first independent public review of commercially available iris recognition algorithms that look for matches within a large database of potential identities.
The 92 algorithms analyzed came from nine private companies and two university labs, which subjected their software to an open competition conducted by NIST, which tested algorithms to identify individuals from a database of iris images representing more than 2.2 million people.
NIST found variations in the accuracy rates of the algorithms examined. No software turned out perfect, but success rates fluctuated between 90 and 99%. Some also had up to ten times more errors than others.
Speed is a factor in iris recognition, especially when looking for matches in a broad data set. NIST found that some of the algorithms could be so fast as to review a dataset with hundreds of millions of records in less than 10 seconds, using a standard computer; however, speed implied limitations in accuracy.
NIST hopes its findings will be useful to policymakers around the world who are interested in incorporating iris recognition and other biometric modalities for identification purposes.
