NIST rates facial recognition systems

Software from three companies has been identified as the most reliable in tests of 14 facial recognition products by the National Institute of Standards.

After testing 14 facial recognition products, the National Institute of Standards has identified software from Cognetic Networks Inc. of Houston, Eyematic of Los Angeles and Identix Inc. of Minnetonka, Minn., as the most reliable.

For its Face Recognition Vendor Test 2002, NIST evaluated facial recognition software by comparing 121,589 images of 37,437 people, an extremely large data set.

The USA Patriot Act of 2001 mandated that NIST do the tests as part of a broader initiative to use biometric systems at border crossings. The agency ran the tests in July and August at the Naval Surface Warfare Center in Dahlgren, Va.

The test had three parts. First, NIST asked the systems to match a facial image against the database of images and find 10, 20 or 25 similar images. Next, the systems had to verify identities using the database of images. Finally, NIST checked each system's reliability under different lighting conditions and monitored the speed of each application.

The three top-rated systems verified identities correctly 87 percent to 90 percent of the time with a false-alarm rate of 1 percent. When NIST specified a false-alarm rate of 0.1 percent, the success rate dropped to between 79 percent and 82 percent.

When checking facial images against a watch list of 25 images at a false-alarm rate of 1 percent, the top three systems were accurate about 80 percent of the time. The success rate fell to below 60 percent when NIST expanded the watch list to 3,000 images at the same false-alarm rate.

Based on the tests, NIST reached some general conclusions about facial recognition systems:

  • The systems recognize men more easily than women.


  • Younger people are harder to recognize than older people.


  • Recognition from video imagery is not much better than from still images.


  • Even topflight software is not sensitive to typical indoor lighting changes.


  • The technology has improved substantially over the past two years.


  • Click to link to the results of the NIST test