FB pixel

NEC America explains top biometric accuracy finish for masked faces in DHS Rally

 

mask detection

NEC Corporation of America’s face biometrics outperformed the median competition by 19 percent for identifying subjects wearing masks in the latest Biometric Technology Rally held by the U.S. Department of Homeland Security (DHS), the company has calculated.

The NEC-submitted algorithm, codenamed ‘Alan’ for the anonymized test, scored a 98.7 percent true identification rate (TIR) when used with the top-performing acquisition system, the best result in the Rally.

“Face masks are not a new challenge to NEC. They have been quite commonly used to protect public health throughout Japan and parts of Asia ever since H1N1 in 2009, or even before that,” explains Kris Ranganath, CTO of NEC Corporation of America. “We’re pleased that this deep experience allows us to respond to the COVID-19 pandemic quickly to address the hygiene issues with effective contactless technology.”

The 2020 DHS Biometric Technology Rally was held at the agency’s Maryland Test Facility (MdTF), as the annual event is each year. Part two of the test was held in January.

NIST testing has suggested significant gains in biometric accuracy with masked faces by algorithms from many companies over the course of the pandemic.

In the DHS Rally, NEC America’s algorithm outperformed its competition in combination with most of the image acquisition systems tested.

The best test of pure algorithm performance is considered to be the ‘focussed’ category, which excludes errors from the front-end capture system, according to the announcement. The company says including these external errors, its algorithm achieved a 95.9 percent TIR, 4 percent better than the second-place algorithm.

The algorithm also exceeded DHS’ criteria of 99 percent or higher matching rate for people without masks. NEC America’s ‘Pearl’ entry topped biometric accuracy in that test, at 99.8 percent TIR without image acquisition errors, and 99.3 percent with errors included.

“This is one of the best opportunities to see how well facial recognition systems work, how components operate together, and how they fare against each other in similar conditions,” says Benji Hutchison, president of federal operations for NEC Corporation of America. “We’re pleased to see that our significant investments in research and development translates to a much better and more reliable customer experience in all sorts of environments.”

Article Topics

 |   |   |   |   |   |   |   |   |   |   |   | 

Latest Biometrics News

 

Foundational ID and biometric binding underpin digital trust

Biometric fundamentals and foundational ID are considerations at the very beginning of the identity lifecycle that are sometimes overlooked. If…

 

ICE withdraws $125M sole-source Thomson Reuters surveillance data deal

U.S. Immigration and Customs Enforcement (ICE) has withdrawn plans to award Thomson Reuters Special Services (TRSS) a five-year, $125 million…

 

TECH5 posts fastest search time in NIST IREX 10 iris evaluation

TECH5 has recorded the fastest search time in the current two-eye developer table of the National Institute of Standards and…

 

OCR Studio updates AI MRZ scanning for difficult remote ID captures

OCR Studio has updated its on-device software to scan the machine readable zones (MRZs) of passports, identity cards, visas, and…

 

Beyond accuracy: What NIST’s latest age estimation results mean for age assurance

The latest results of the National Institute for Standards and Technology (NIST)’s Face Analysis Technology Evaluation Age Estimation & Verification…

 

Latest round of BixeLab biometrics evaluations include Level 3 PAD, IAD tests

A pair of facial recognition providers have passed Level 3 independent biometric Presentation Attack Detection (PAD) and Injection Attack Detection…

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