Innovatrics, ROC improve rankings in NIST ELFT, rising to 2 and 3 respectively

Innovatrics is celebrating success in the latest National Institute of Standards and Technology (NIST) Evaluation of Latent Fingerprint Technologies (ELFT) benchmark for fingerprint biometrics. A release says the result for Innovatrics’ latest submission, 0010, updated on February 2, “confirms its latent fingerprint algorithm among the most accurate in the world for false negative identification rate (FNIR) at rank ≤1.”

Testing for ELFT aims to “evaluate the current state of the art in latent feature-based matching by comparing the accuracy of searches using images alone with searches using different feature sets.” Recent ELFT benchmarks have shown continuing gains in the accuracy and speed of the top-performing latent fingerprint biometrics matching algorithms.

Innovatrics’ result improves on its last submission from May 2025, climbing two places in the overall rankings.

For a dataset provided by the FBI, consisting of 285 operational probes collected from a particular type of crime, Innovatrics scored 98.6 percent in rank-1 hit rate – a Cumulative Match Characteristic (CMC) metric derived from FNIR by computing 100 × (1 – FNIR), and among the most widely used methods by fingerprint laboratories in which results are reviewed manually. According to the NIST nutrition report for Innovatrics IEngine Matcher, that puts it at number two overall.

The firm says being among the best in the rank-1 category is “particularly important because it makes the lives of examiners easier and reduces time spent on manual comparison by quickly identifying the correct match.”

For another dataset, consisting of 5,259 probes collected operationally by the United States Department of Defense, Innovatrics has “remained as one of the most accurate algorithms in both rank and threshold-based search.”

“Innovatrics latest achievement highlights the exceptional quality of its algorithms, especially given that they currently outperform industry leaders and well-established competitors,” says the company’s release.

ROC, Griaule among companies for recent ELFT testing

Right behind Innovatrics for rank-1 hit rate is U.S. firm ROC, which posted a rank-1 hit rate of 98.2 percent for the January 2026 submission of its 0017 algorithm. It moves up one place to number 3. Also tested in January was Brazil’s Griaule Biometrics, which posted a rank-1 hit rate of 85.3 percent for its 100F algorithm.

The highest value recorded for rank-1 hit rate is 98.9 percent, for Beijing Hisign Technology Co, Ltd.’s 001B: HisignImplementation Matcher 2.2, submitted in April 2025.

Article Topics

biometric testing  |  biometrics  |  ELFT  |  fingerprint biometrics  |  fingerprint recognition  |  Griaule Biometrics  |  Innovatrics  |  NIST  |  ROC