According to a new report by the National Institute of Standards and Technology (NIST), the speed with which iris recognition systems can scan a person’s eyeball for identification have become much faster. However, much work remains to be done when it comes to improving the accuracy of these scans.
The Iris Exchange (IREX) III report, which evaluated 92 different iris recognition algorithms from nine private companies and two university labs, found that accuracy of iris scans varied widely across the algorithms tested. The aforementioned companies and labs submitted their software to an open competition held by NIST, which sought to identify individuals from an eye image database consisting of more than 2.2 million people.
According to a statement, success rates reportedly ranged between 90 and 99 percent among the algorithms and some produced as many as 10 times more errors than others. In addition, while the tests found that some algorithms were fast enough to run through a dataset equivalent of the size of the entire U.S. population in less than 10 seconds on a typical computer, there could be accuracy limitations.
To improve the accuracy of iris recognition, a related NIST report suggested that operators control image collection more tightly, which would result in higher quality iris images.
"When combined with the feedback that this study provides to the industry and the use of the iris in combination with other biometrics, the findings will push accuracy toward 100 percent," said Patrick Grother, a scientist in NIST's Information Access Division.
For links to the reports, visit http://iris.nist.gov/irex.
