Iris-recognition biometric scans used at many security checkpoints may be less reliable than previously thought, researchers at the University of Notre Dame have found. In a paper to be presented next week at the IEEE Computer Vision and Pattern Recognition conference, two scientists claim that patterns in the irises change significantly as people age. The study used commercial iris matching software to compare 20,000 images of 644 irises, using iris images taken between one month and three years apart.
When the researchers compared photos taken a month apart, they found few instances where the system didn’t match two irises of the same person. However, as the time between photos increased, the false mismatch rate increased: when iris photos taken three years apart were compared, the false mismatch rate was 153% higher than that of photos taken a month apart. In practical terms, this still means that only about 2.5 out of 2 million iris scans will be mismatched after three years. Over time, however, the effect will be compounded, says co-author Kevin Bowyer, potentially locking some people out of systems or letting others cheat security checkpoints. This means that at the very least, the images need to be updated every few years, and future pattern recognition systems may need to account for changing irises as well as different lighting conditions and other factors.