Eye-tracking tech helps CBP detect imposters
- By Susan Miller
- May 04, 2017
Nearly one million people are screened every day by Customs and Border Protection agents who secure and manage 328 ports of entry across the country.
A critical part of that screening is verifying the identities of those entering the United States, and frontline operators must be able to accurately and efficiently spot imposters. Currently, agents are trained through classroom instruction, PowerPoint presentations and training booklets, but those tools do not give trainers or trainees any feedback on the trainees' visual search patterns.
But thanks to a collaboration among the Department of Homeland Security’s Science and Technology Directorate, CBP and the Federal Law Enforcement Training Center (FLETC), CBP has developed a technology to optimize imposter detection training. Called Eye-dentify, the tool tracks trainees’ eye movements during training so instructors can provide immediate feedback on performance.
Eye-dentify is built on the technology in DHS’s ScreenADAPT, which uses eye-tracking technology to examine visual search performance of Transportation Security Agency staff as they monitor X-ray images of bags boarding commercial aircraft.
Similarly, Eye-dentify tracks agents’ eye movements during training to analyze how they examine an ID or face. Before Eye-dentify, an instructor had no way of knowing where trainees were looking or if they were following the procedures. Eye-dentify makes previously unobservable performance now observable, DHS said.
When using Eye‑dentify, trainees are given immediate feedback that includes not only whether their decision was correct, but how the face or image was scanned. Officers can then review their performance and “identify” the facial features they did or did not sufficiently examine. Trainers and trainees can review the accuracy and eye-tracking data for individuals or get aggregate data for an entire class.
“The training we conduct for imposter detection does not include the ability to collect the objective, immediate feedback on trainee performance. We need that feedback to address deficiencies in the limited class time we have with students,” said CBP’s Basic Training Branch Chief Michael Shekot. “Eye‑dentify allows us to maximize the effectiveness of our time with each student … to see how someone is performing their inspection and offer feedback that addresses the root cause of errors or inefficiencies.”
A training effectiveness evaluation across three new hire classes of CBP officers demonstrated a 15 percent improvement in imposter detection. Trainees using Eye-dentify were able to both examine more image pairs in the same amount of time and detect more imposters, DHS said. CBP is currently using eight Eye‑dentify systems in its FLETC classroom, and DHS S&T will transition another four systems in July.
The technology will soon be used for training for other tasks as well, such as searching for signs of fake or altered IDs/credentials, pre-assault indicators or suspicious behavior.
More information on Eye-dentify can be found on this Facebook Tech Talk.
Susan Miller is executive editor at GCN.
Over a career spent in tech media, Miller has worked in editorial, print production and online, starting on the copy desk at IDG’s ComputerWorld, moving to print production for Federal Computer Week and later helping launch websites and email newsletter delivery for FCW. After a turn at Virginia’s Center for Innovative Technology, where she worked to promote technology-based economic development, she rejoined what was to become 1105 Media in 2004, eventually managing content and production for all the company's government-focused websites. Miller shifted back to editorial in 2012, when she began working with GCN.
Miller has a BA and MA from West Chester University and did Ph.D. work in English at the University of Delaware.
Connect with Susan at [email protected] or @sjaymiller.