Skip to main navigationSkip to main content
The University of Southampton

Multiple types of threat objects in airport baggage security scans can make those threats difficult to spot.

Published: 12 November 2015
Airport security

Research finds that multiple types of threat objects in airport baggage security scans can make those threats difficult to spot.


Searching for explosives and guns in people’s luggage is an essential part of safeguarding our national security – but this study reveals that having to simultaneously search for two types of targets – metal weapons and explosives – can lead to errors. X-ray scans are difficult to interpret because they involve images of overlapping transparent objects, making those objects difficult to recognise.


Psychologist at Southampton, Dr Tamaryn Menneer, comments: “Our results suggest that airport security might be improved if instead of having one screener at each point – trained to look for metal weapons and explosives – there were instead two screeners, one specialising in metal weapons and the other in explosives.”


The study is being presented as part of the Economic and Social Research Council’s Festival of Social Science, which takes place from 7-14 November 2015 with over 200 free events nationwide – including Southampton.


Dr Menneer and colleagues from the University of Southampton and the University of Massachusetts asked students to look through over 5,000 X-ray images, searching for either metal weapons like guns and knives, which show up as blue in X-ray images, or improvised explosive devices, which show up as orange. Some of the students were asked to search for both objects at the same time.


The study showed that looking for two targets simultaneously reduced accuracy and lengthened reaction times. Students took longer to make a decision, failed to spot targets when they were present and incorrectly thought targets were there when they weren’t. Although the students’ overall performance improved as they gained more practice at interpreting X-ray scans, those tasked with looking for two objects still consistently made more errors.


In order to find out what was making the students less accurate, and provide an insight into the cognitive processes going on in the brain, each person had their eye movements tracked as they searched.


“What we found was that when students were looking for just one object they would fixate on objects that were a similar colour to that target,” says Dr Menneer. “In other words, they looked at blue objects if they were looking for metal weapons and orange objects if they were looking for explosives. However when they were searching for two different types of objects at the same time they were much less precise in what they chose to look at. They looked proportionally less at both blue and orange objects and more at the irrelevant green items.”


Spotting weapons in X-ray scans is a complex task because although your luggage is 3D, the image isn’t, which means all the objects in the suitcase are superimposed on top of one another. This means that screeners can’t use information about depth to help recognise separate objects; a limitation that might decrease recognition and therefore reduce accuracy.


Dr Menneer and colleagues have teamed up with computer scientist Professor Nick Holliman at Newcastle University to use X-ray images that are split into multiple depth planes, allowing the viewer to see which objects are in front and which are behind one another. It’s hoped that this extra information about depth will help people learn how X-ray images of individual objects combine together when they overlap, and therefore train people to interpret 2D images more easily.


“The human brain has evolved to interpret and analyse the world around us, which is solid and in 3D,” explains Dr Menneer. “It’s used to getting depth cues to help interpret how big objects are and how they overlap with others, and to help us recognise objects in the real word. X-ray images are difficult to interpret because the brain isn’t generally experienced with processing overlapping transparent items in which you can see through one object to the next. We hope that the training in 3D will help people learn to interpret 2D images more efficiently.”


Dr Tamaryn Menneer’s research will be explained in a hands-on exhibit at a local school. The University of Southampton is holding three other events as part of the festival:


- ‘Behind the walls of your local nick’ – a chance to hear from HMP Winchester’s Governor in Charge, David Rogers – with Lecturer in Criminology, Dr Alisa Stevens.


- ‘Observing Southampton through the web’ – examines the Southampton area using real time and historic data of online digital traces of web users.


- The problem with ‘immigration street’: debunking myths – a private event aimed at sixth form students, to encourage them to debate the impact of immigration in the UK.



Privacy Settings