DHS to hold facial recognition rally in fall

To better manage the country’s ports of entry, the Department of Homeland Security is looking for companies with state-of-the-art facial recognition capabilities.

The DHS Science and Technology Directorate is planning its third annual year Biotechnology rally in September to evaluate the latest advances in facial recognition. The agency, as well as volunteers and industry stakeholders, will examine how the company’s products apply to various environments and conditions that have foiled biometric technology in the past.

At the rally, which will be held at the Maryland Test Facility in Upper Marlboro, Maryland, dealers will showcase their products to both the DHS and the public.

Arun Vemury is the director of S & T’s Biometric and Identity Technology Center.

“The biometric rally is an opportunity for S&T to get in direct contact with the industry,” said Arun Vemury, director of S & T’s Biometric and Identity Technology Center. Federal Drive with Tom Temin. “We work with them to make sure they better understand the common DHS use cases, talk more about performance, and really focus on empowering them to differentiate their product and technology and delivering them tangible Feedback so they can go back and improve the technology. “

A diverse set of around 600 volunteers will take part in the rally to test the performance of the presented matching algorithms. S&T – which publishes some international standards for evaluating the performance of biometric technologies – will then work with cross-agency partners to evaluate facial recognition capabilities.

The DHS will share the feedback with the stakeholders of the authorities and international partners who are also involved in border security and control.

Current challenges in face recognition include identifying twins and people who wear respirators. But the agencies, in addition to their private sector counterparts, are working to minimize these shortcomings.

“We are trying to optimize the entire system, the entire process, to reduce errors and help people get through the screening process quickly,” said Vemury.

Groups like the National Institute of Standard and Technology have performed research with matching algorithms for twins. And in 2020 DHS Tested facial recognition performance for mask wearers. The agency found that the top performing camera and algorithm combination gave an accuracy rate of 96% – a percentage that suggests the need for further testing.

“The results so far have been surprisingly promising and good,” said Vemury. “While there are some systems that work extremely well, there is still a lot of room for improvement across the board.”

However, challenges remain with the performance of facial recognition tools in various environments, e.g. The technology must also be able to recognize different genders, races and other demographics.

According to Vemury, S&T is urging these companies to target their products to a wide range of performance characteristics rather than specific applications so that DHS can apply the facial recognition tools to different missions.

“We’re not just looking at a DHS component – we’re looking at DHS components and missions and abstracting out the technical parts,” said Vemury.

Comments are closed.