DHS S&T Biometric Technology Rally results suggest face best for fast processing

Four matching systems produced 100 percent true identification rates on the same acquisition system during the Biometric Technology Rally held recently by the Department of Homeland Security Science and Technology Directorate (S&T) at its Maryland Test Facility.

DHS S&T tested 14 acquisition systems and 15 matching algorithms, using iris, face, and fingerprint recognition, alone or in combinations, to test the effectiveness, efficiency, and user satisfaction of systems designed to process transactions in under 10 seconds. The results, shown with aliased company names at the Maryland Test Facility’s website, shows systems generally performed well in terms of throughput and satisfaction, but biometric performance ranged from well below the Rally threshold to perfect during the trial. None of the systems using iris or fingerprint biometrics achieved the threshold, while several facial recognition systems did.

The acquisition system that four algorithms returned 100 percent true identification rates on was Tascent, which was code-named Teton. One of the four algorithms with the high matching score was Rank One, under the name “Wabash.”

Arun Vemury, Director of DHS S&T’s Biometric and Identity Technology Center (BI-TC) compared the Rally project to the children’s book ‘Where’s Waldo,’ but scaled up as high as millions of people.

“We found that several use cases, such as border security, aviation security, and physical access controls at government facilities have similarities. The intent here is to continue to push technology developers to continue to improve technologies, make them easier, faster, and more accurate,” said Vemury in an announcement. “We are working with technology providers to clarify our expectations that the technology work more effectively for all users, reduce errors, make systems more cost-effective for governments, and offer a better user experience for the people undergoing the screening process.”

The 2019 Rally was the second one, and DHS S&T built on its experience in the inaugural Rally last year to carry out a robust and repeatable testing process, which included collecting data for biometric capture error rates, collection times, matching error rates, and user satisfaction comparing results across all systems, according to the announcement. The feedback gathered is meant to enable the industry to improve screening capabilities and inform the next generation of biometric technology.

“When you conduct screening operations, you usually have an officer involved. If you have a ratio of one officer to one person, that’s very time consuming and expensive. These technologies, the way they were tested, no one was providing instruction to the user, so operators can focus on higher value security tasks. It was supposed to be intuitive and easy to someone who wants to participate but may not be familiar with the new process,” comments Vemury.

“We encouraged industry to consider creative solutions with few constraints on how they use their dedicated 6’x8’ space. But, we tried to get people through the entire process in less than 5 seconds, maybe up to ten seconds. It really has to be fast and very effective for all users.”

DHS still has some political or social acceptance barriers, in addition to technical challenges, to its Biometric Entry/Exit and other department’ agency programs.

Article Topics

accuracy  |  biometrics  |  border management  |  data collection  |  DHS  |  facial recognition  |  fingerprint recognition  |  iris recognition  |  testing