Sci-fi films routinely lead viewers to believe that scanning an individual's iris is a proven way to identify them, but in practice, the results haven't always been 100% dependable. One of the most significant challenges isn't the technology, but how slight changes in the structure of the iris can throw off calculations used in comparing images of the human eye.
The long-term stability of the iris' distinguishing characteristics, critical for biometric identification, had come under question when a recent study of several hundred subjects found that iris recognition becomes increasingly difficult over a period of three years, consistent with an aging effect.
The latest in an ongoing series of studies of iris recognition for biometric identification, however, refutes that. Scientists at the National Institute of Standards and Technology (NIST) have found that the unique characteristics of the iris in the average person do not change for at least nine years. The results of the study, conducted by researchers in NIST's Information Access division, suggest that iris recognition of average individuals will remain viable for decades. They also imply that identity program managers may not need to recapture iris images as frequently, which factors into the total overall cost of maintaining iris recognition systems.
The new study by NIST researchers used two large operational data sets, including one of nearly 8,000 recurrent travelers across the Canadian-American border, involving millions of images. The travelers, like the woman pictured here in a photograph supplied by the Canadian Border Services Agency, use an iris identification system to confirm the individuals' identity. The system is part of a joint Canadian and American program to help people move quickly across the border. The study examined images that had been captured at least four years and up to nine years previously. NIST researchers found no evidence of a widespread aging effect.
NIST has been working with a variety of organizations to help improve the use of iris recognition systems. In that vein, it established the Iris Exchange program in 2008. The program has sought to establish standards for iris recognition, as well as the development and deployment of systems used to capture and identify iris images. Sponsors of the program include the FBI's Criminal Justice Information System Division and the Office of Biometric Identity Management in the Department of Homeland Security.