New crop of latent fingerprint algorithms set new highs in NIST testing

The U.S. National Institute of Standards and Technology has completed eight evaluations of biometric algorithms for latent fingerprint matching since the start of this year, which set new highs in accuracy and speed.

NIST has tested new submissions from Hisign, Dermalog, ROC, Neurotechnology, Griaule, Idemia, Peking University and Innovatrics since January 24, 2024 for its Evaluation of Latent Fingerprint Technologies (ELFT).

Idemia scored the lowest false positive identification rate (FPIR), at just under 8 percent. Hisign was next at 11 percent. ELFT assessments are carried out with the false non-identification rate (FNIR) set at 1 percent. ROC had the fastest mean mated search duration, at 15 seconds, over a hundred times faster than the next-fastest algorithm.

The top rank-1 hit rate was 96.5 percent, from Idemia and Hisign.

Idemia says in an announcement that its result is the fourth consecutive time it has achieved the best accuracy and the top speed among the most accurate algorithms.

The algorithm submitted by Idemia is already available in its flagship Multi-Biometric Search Services (MBSS), according to an announcement from Idemia Public Security, the group’s biometric solutions division.

Idemia also scored high marks for fingerprint biometrics accuracy in the PFT III just before the calendar flipped to 2024.

Article Topics

algorithms  |  biometric testing  |  biometrics  |  DERMALOG  |  ELFT  |  fingerprint biometrics  |  Griaule Biometrics  |  Hisign  |  IDEMIA  |  Innovatrics  |  Neurotechnology  |  NIST  |  ROC