IDEMIA Performs Well in NIST’s Airport Simulation

IDEMIA’s facial recognition algorithm continues to perform well in independent NIST testing. The company retained the top spot that it achieved in the 1:N matching category earlier this year, and supplemented that with a top-two performance in the Face Recognition Vendor Test (FRVT) for Identification for Paperless Travel and Immigration.

As the name would suggest, the latter test focused specifically on travel applications of facial recognition technology. The NIST set out to evaluate the performance of different algorithms in an airport setting, and created a simulated flight departures environment in order to do so. From there, the agency used a gallery of 420 arriving faces to gauge each algorithm’s ability to match people’s faces at passenger screening checkpoints. IDEMIA placed second in that regard, and did not return a single false negative after 536 aircraft boarding exercises.

“NIST’s results confirm the robustness of our technologies with regard to managing different demographics,” said IDEMIA CTO Jean-Christophe Fondeur. “IDEMIA’s facial recognition technology achieves the most accurate results and delivers a key competitive advantage when handling complex scenarios.”

According to Fondeur, the strong results should make IDEMIA’s solution an appealing option for airports that are looking to update their passenger processing procedures. IDEMIA’s facial recognition technology has already been deployed at airports and border checkpoints all over the world, with Los Angeles International Airport and the Santos Dumont airport in Rio de Janeiro standing as two of IDEMIA’s most recent projects.

The company also integrated its facial recognition tech into a new TravelKiosk that debuted in October. That Kiosk will be used to expedite border screenings in Iceland and France, and can be combined with IDEMIA’s new Digital Travel Credential to further streamline the passenger experience. That credential links a traveler’s facial biometrics to a secure digital credential that can be stored on a standard smartphone.

–

August 9, 2021 – by Eric Weiss