Fingerprint ID systems have the edge, NIST finds
The most accurate fingerprint systems far outperform the most accurate face recognition systems, the National Institute of Standards and Technology recently reported at <a href= "http://fpvte.nist.gov">fpvte.nist.gov</a>.
The most accurate fingerprint systems far outperform the most accurate face recognition systems, the National Institute of Standards and Technology recently reported at fpvte.nist.gov.
The top performers for fingerprints, with consistent accuracy rates higher than 98 percent, were systems from NEC Technologies Inc. of Itasca, Ill.; Sagem Morpho Inc. of Tacoma, Wash.; and Cogent Systems Inc. of Alhambra, Calif.
NIST conducted the Fingerprint Vendor Technology Evaluation 2003 last October and November. Under the USA Patriot Act, the agency must certify biometric technologies that could be used in the U.S. Visitor and Immigrant Status Indicator Technology program. NIST conducted similar tests on facial-recognition systems in 2002.
The tests of 34 systems from 18 companies used more than 48,000 sets of prints supplied by federal and state sources. Vendor personnel configured and operated their own systems for the trials at NIST in Gaithersburg, Md.
Testing covered one- and two-finger print sets, as well as prints from four or more fingers. The number of fingers and quality of prints made a significant difference in accuracy.
With 0.01 percent false positives, the NEC system had 98.6 percent accuracy for single-finger prints, 99.6 percent for two-finger prints, and 99.9 percent for four-, eight- or 10-print sets. The Sagem and Cogent systems had slightly lower accuracy rates around 95 percent.
In contrast, the best facial-recognition systems tested in 2002 had 71.5 percent accuracy with a false-positive rate of 0.01 percent.
Not all the fingerprinting systems were highly accurate, however. NIST reported a large performance gap between the best and the rest.
Although print quality made a big difference, the type of scanner or acquisition device that gathered the prints did not necessarily determine the quality of the results. NIST did not evaluate reliability and did not consider cost-effectiveness or latent fingerprint identification.
NEXT STORY: Malicious code targeting PDAs identified