The accuracy of facial-recognition algorithms has improved over the past three years, according to a U.S. National Institute of Standards and Technology (NIST) study. The study showed a 10% to 30% improvement from 2010 to 2013 in the accuracy of 50 facial-recognition algorithms.
The best algorithm tested in 2013 had a failure rate of 4.1%, down from 5.7% in 2010. The results demonstrate facial-recognition technology will prove to be more useful to users in the future.
The researchers also found that rates of missing facial matches increase as the database grows. "This fact is largely responsible for the usefulness of face-identification algorithms in the marketplace, such as at DMV settings," says NIST researcher Patrick Grother.
The best results occur when the image being compared with those in the database are of good quality, the researchers note.
The main challenges Grother says organizations must tackle to effectively use facial-recognition software for authentication are having developers create algorithms that produce better recognition of lower quality images and getting systems owners to collect better quality photographs.
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found