FB pixel

Updated Innovatrics ABIS algorithm takes back rank 1 in latent accuracy evaluation

Updated Innovatrics ABIS algorithm takes back rank 1 in latent accuracy evaluation
 

An updated biometric algorithm for latent fingerprint identification from Innovatrics has landed at the top of the U.S. government’s Evaluation of Latent Fingerprint Technologies (ELFT).

The algorithm submitted on December 11 to the National Institute of Standards and Technology’s (NIST’s) ELFT scored the top rank-1 hit rate at 98.2 percent, and also placed first in benefit from EFS features and reviewing 5 ranks.

That high rank-1 hit rate was achieved in comparisons to the FBI-provided solved dataset #1, which consists of 516 probe images. The company scored a false non-identification rate (FNIR) of 0.0659 at false positive identification rate (FPIR) of 0.01 for the same dataset, which trails only the algorithm from Neurotechnology evaluated weeks earlier.

In searches against DoD Dataset #1, the largest in the test with 5,259 latent fingerprint samples, Innovatrics delivered the highest accuracy in terms of rank and searches to a threshold.

“Our latest achievement in the NIST ELFT evaluation underscores Innovatrics’ unwavering commitment to excellence in biometrics,” says Innovatrics ABIS Business Unit Head Matus Kapusta. “In a world where secure and reliable identification is more critical than ever, our cutting-edge solutions empower businesses, governments and law enforcement to meet the highest standards of accuracy and efficiency, ensuring trust and safety across global applications.”

Innovatrics has consistently kept near the top of the ELFT leaderboard, and also landed in the top 10 for facial recognition accuracy on the FRTE 1:N leaderboard in a November update.

Remote onboarding primer published

Innovatrics has also published the final chapter in its review of remote user onboarding. “Control the Onboarding Flow: Customizing Your Digital Identity Solution” describes the benefits of its Digital Identity Service, the importance of deduplication and why businesses should set their own confidence thresholds.

The post is one in a four-part series on remote onboarding. The other three entries in the “Elements of Digital Onboarding” address the sectors making use of remote identity verification and the technologies they use, how face biometrics tie in with liveness detection and other tools and the details of ID card scanning.

Related Posts

Article Topics

 |   |   |   |   |   |   |   | 

Latest Biometrics News

 

Biometrics consolidation reflects shift to continuous identity

AI-driven fraud is reshaping the biometrics industry. This week’s acquisitions, product launches and market developments all point toward the same…

 

Australia’s national face matching system expands with NSW participation

New South Wales (NSW) is joining Australia’s national face matching network, allowing driver licence and photo card images to be…

 

From behavior to payment: Visa builds trust across the customer lifecycle

Visa’s agreement to acquire BioCatch for $2.4 billion raises an important strategic question: Why would a global payments network invest…

 

Why deepfake detection is becoming trust infrastructure

For many, the word “deepfake” still has a twinge of novelty about it – a hint of science fiction. But…

 

Brazil age assurance challenge overlooks privacy-preserving options

Where there are online safety laws, there is litigation aiming to curb them. So it goes in Brazil, where the…

 

Post-quantum identity expands beyond cryptography

As governments and enterprises prepare for practical quantum computing, the focus has largely been on post-quantum cryptography. IBM is targeting…

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Market Intelligence

Featured Company

Biometric Update Podcast

Most Read This Week

White Papers

Latest Webinars

Biometrics Industry Events