The U.S. National Institute of Standards and Technology (NIST) has officially relaunched its fingerprint biometrics matching evaluations after more than a decade, marking a significant milestone for the global biometrics industry. The initiative resumes 13 years after the last such large-scale testing and is set to provide updated benchmarks for one-to-many fingerprint identification algorithms used worldwide.
The program, previously known as the Fingerprint Vendor Technology Evaluation (FpVTE), has been reintroduced under a new name: Friction Ridge Image and Features (FRIF) Technology Evaluation Exemplar One-to-Many (E1N). Its purpose remains to rigorously test and measure the accuracy, reliability, and efficiency of fingerprint matching algorithms that are integral to civil identity programs, passports, elections, and large-scale automated biometric identification systems (ABIS).
So far, only two companies — Tech5 and Innovatrics — have submitted their algorithms for testing, signaling the early phase of the evaluation process. NIST is using three separate datasets to assess the performance of these algorithms. These include Class A, made up of index fingers only; Class B, consisting of “4-4-2” identification flat captures; and Class C, which incorporates full ten-finger datasets.
Initial results show Tech5 taking the lead in performance across multiple benchmarks. Against Class A’s 1.6 million images, Tech5’s algorithm achieved a false non-identification rate (FNIR) of 0.001 at a false positive identification rate (FPIR) of 0.001 or lower, with a Rank-1 FNIR of 0.0004. For Class C datasets, Tech5 has recorded the best outcomes across plain-plain, plain-rolled, and rolled-rolled matchings, with FNIR results as low as 0.0003 in rolled-rolled comparisons. The company also demonstrated strong results in template creation speed and footprint efficiency, making its algorithm well-suited for deployment in high-volume identification systems.
Innovatrics, another early participant, also posted competitive results. At FPIR 0.001 or lower, the company’s algorithm achieved FNIRs of 0.0160 for left slaps, 0.0120 for right slaps, 0.0022 for combined slaps, and 0.0011 for identification flats. Its Rank-1 FNIR results also showed promising accuracy, ranging from 0.0092 to as low as 0.0005 in certain categories.
The relaunch of these evaluations is seen as a critical development for governments and organizations procuring ABIS solutions, as they now have updated, independently verified benchmarks to reference. The last similar NIST evaluation, conducted in 2012, attracted submissions from major global players, including NEC, Morpho, Dermalog, and BIO-key.
Speaking about the importance of the evaluation, Ameya Bhagwat, Chief Revenue Officer of Tech5, said, “This evaluation was highly needed, because the previous similar 1:N fingerprint NIST testing was conducted 13 years ago under FpVTE. The end customers are looking at these tests as a reference when selecting large-scale ABIS platforms for their projects in civil identity, foundational identity, elections, passport systems and the like. We at TECH5 are very proud that this fingerprint algorithm, that is used in the T5-OmniMatch ABIS platform, shows the best results.”
NIST clarified that the FRIF Technology Evaluation focuses specifically on template creation and search algorithms within ABIS platforms rather than evaluating entire systems. This distinction ensures the results are narrowly tailored to measure algorithmic capabilities, providing objective data for stakeholders deploying biometrics solutions at scale.
With the biometric identification market expanding rapidly and global reliance on digital identity verification increasing, the relaunch of NIST’s evaluation program is expected to set the stage for new innovations and competitive benchmarks in the fingerprint biometrics space.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem
