Take a sneak peek at the new NIST.gov and let us know what you think!
(Please note: some content may not be complete on the beta site.).
NIST Authors in Bold
|Author(s):||Charles L. Wilson; Patrick J. Grother; Ross J. Micheals; S C. Otto; Craig I. Watson; R. A. Hicklin; H Korves; Brad Ulery; M Zoepfl;|
|Title:||Fingerprint Vendor Technology Evaluation 2003: Summary of Results and Analysis Report|
|Published:||June 30, 2004|
|Abstract:||The Fingerprint Vendor Technology Evaluation (FpVTE) 2003 was conducted to evaluate the accuracy of fingerprint matching, identification, and verification systems. The FpVTE is one of the tests that NIST has conducted in order to fulfill part of its PATRIOT Act mandate. Additional evaluations include the testing of the FBI IAFIS system, the US-VISIT IDENT system and SDKs (Software Development Kits) from several vendors. Eighteen different companies competed in FpVTE, and 34 systems were evaluated. Different subtests measured accuracy for various numbers and types of fingerprints, using operational fingerprint date from a variety of U.S.Government sources. The most accurate systems were found to have consistently very low error rates across a variety of data sets. The variables that had the clearest effect on system accuracy were the number of fingers used and fingerprint quality. An increased number of fingers resulted in higher accuracy: the accuracy of searches using four or more fingers was better than the accuracy of two-finger searches, which was better than the accuracy of single-finger searches. The test also shows that the most accurate fingerprint systems are more accurate than the most accurate facial recognition systems, even when comparing the performance of operational quality single fingerprint to good quality face images.|
|Citation:||NIST Interagency/Internal Report (NISTIR) - 7123|
|Keywords:||fingerprint accuracy,fingerprint matching,number of fingers,vendor evaluation|
|PDF version:||Click here to retrieve PDF version of paper (3MB)|