The Department of Homeland Security Science and Technology Directorate has released findings from the 2022 Biometric Technology Rally, an annual event that tested and compared facial recognition systems.
Results proved group biometric processing effective, and showed that the camera and not the algorithm was usually the source of error, if any, DHS S&T said Thursday.
The rally effort was launched in 2018 in an aim to improve national security and find more user-friendly equipment for security checkpoints. The 2022 rally sought technologies that allow small groups of travelers to undergo biometrics processing with accuracy and speed, despite the presence of bystanders in a crowded setting.
Over the course of of 11 days, the event utilized four face recognition systems matched with 10 algorithms. S&T’s Biometric and Identity Technology Center tested 575 participants from 54 countries.
The researchers found that the best system-algorithm combination correctly identified 97 percent of individuals in less than two seconds when processing a group of two or four people. The systems only erroneously captured less than 1 percent of non-users or bystanders.
“The purpose of Rally 2022 was…to further challenge industry to develop technologies that address operational challenges and constraints, while also addressing privacy and civil rights and civil liberties concerns,” said Arun Vemury, senior advisor of the Biometric and Identity Technology Center. “Biometric technologies are continuing to improve in terms of accuracy, speed, and ease of use, so it is important that we ensure these capabilities also include privacy safeguards, including the option to opt out of biometric screening.”
