Neurotechnology hits high score in NIST Proprietary Fingerprint Template evaluation

Neurotechnology claims to have kept its top score in the National Institute of Standards and Technology (NIST) Proprietary Fingerprint Template (PFT) III biometric performance evaluation results, continuing its algorithmic prowess for fingerprint templates from its 2019 submission.

The evaluation results that were released on March 2022 examine the capabilities of Neurotechnology’s biometric algorithm for one-to-one fingerprint verification and the accuracy of final stage fingerprint matchers used in one-to-many Automated Fingerprint Identification System (AFIS) searches against proprietary templates. It folds in PFT II testing, which is a one-to-one fingerprint verification evaluation that looks at categories like plain versus plain, plain versus rolled, and template sizes; and PFT III, a continuation of previous PFT evaluations but with new datasets. Neurotechnology says the March evaluation also tested for PFT 2003 standards.

To assess Neurotechnology’s MegaMatcher algorithms, 22 vendors submitted 39 fingerprint recognition systems to the NIST that were tested 33 times to fingerprint databases held by law enforcement and border agencies. The combined datasets of the Arizona Department of Public Safety (AZDPS) and the Los Angeles County Sheriff’s Department (LASD) conducted tests one to nine for plain-to-plain, 10 to 18 for plain-to-rolled, and 19 to 27 for rolled-to-rolled fingerprint matching. The Dataset of Department of Homeland Security (DHS2) judged experiments 28 to 30 for plain-to-plain fingerprint matching. The combined biometric datasets of U.S. VISIT Point of Entry Data with Bio-Visa Applications (POE+BVA) were used to perform plain-to-plain fingerprint matching tests.

Its performance in false non-match rate at false match rate was the best among the submitters, with the lowest detection error trade-off curves for all fingers, according to the announcement.

However, the average template creation time was 228.2 milliseconds with no failures, contrasted to 204 milliseconds in 2019 and the mean template comparison time was 4.6 milliseconds against 3.7 milliseconds in 2019, showing slightly reduced speed.

Evaldas Borcovas, biometric research lead at Neurotechnology, says about the evaluations, “It is challenging to continuously improve your technology when it is already outperforming all others in the industry, but our team has continued to demonstrate a spirit of innovation, and we are thrilled to see the latest algorithm improve our leading scores even further. NIST is regarded as the most influential and respected organization evaluating biometric algorithms, and their highest evaluation of our technologies shows that our fingerprint algorithm is the most reliable and accurate algorithm available.”

In 2019, the Lithuanian company took the crown for MINEX III (Minutiae Interoperability Exchange) evaluations for fingerprint biometric algorithms. More recently in February 2022, Neurotechnology recorded significant improvements for its face biometrics in both of NIST’s main FRVT evaluations.

Article Topics

accuracy  |  algorithms  |  biometric identification  |  biometric matching  |  biometric testing  |  biometrics  |  fingerprint recognition  |  Neurotechnology  |  NIST  |  PFT III