"Guilt" detector to catching smugglers

Published 8 October 2008

Researchers are looking to increase security at border crossings by developing a computer system that can detect guilt

Dr. Hassan Ugail, head of Visual Computing Research at the University of Bradford’s School of Informatics, is part of a team on a £500,000 project to develop technologies to help the border control agencies in identifying people trying to smuggle contraband goods or narcotics through customs. The project, which starts in December 2008 and will last over two years, also involves Dr. Reyer Zwiggelaar from the University of Aberystwyth and is being funded by the Engineering and Physical Sciences Research Council (EPSRC). The research team will also collaborate with the U.K. Home Office, HM Revenue & Customs, and leading international defence and security technology company QinetiQ.

The aim of the project is to develop technology that can profile people as they pass through border controls to help security agencies identify smugglers. If implemented, the ideal outcome would be to increase identification of smugglers and decrease the amount of contraband and drugs entering the United Kingdom.

The technology will use a technique called “real-time dynamic passive profiling” based on the modelling of facial expressions, eye movement, and pupil changes in both the visual and thermal terms. This will then link to malicious intent and physiological processes such as blood flow, eye movement patterns and pupil dilation. Ugail explains:

What we are proposing to develop is essentially a passive lie detector. We aim to automatically analyse peoples’ facial expressions and eye movements in response to a series of questions through video images and computer-based intelligent algorithms. For example, trained officers at the border control points are very good in spotting people carrying contraband by simply analysing their facial expressions in response to questions but it is tricky to teach a machine to do this.

However, some smugglers can be very good actors who can easily hide their emotions. That’s why we aim to extend this study to other non-visual domains such as the use of thermal imaging to study facial blood flow which is extremely hard to control.

If successful, this work has a potential beyond border control applications. For example, the system could be used for police interrogations and interview scenarios. “Who knows — it could even be used to enhance our real-time computer gaming experiences,” adds Ugail.