Quicker baggage scans for safer flights

University of Manchester mathematicians are combining 2D X-rays from airport hold baggage scans into 3D images to better detect dangerous content and help reduce bottlenecks.

Most airport X-ray machines currently in use only provide a 2D view of a bag’s contents. The new system, now at the working prototype stage, not only generates better images but also does so very quickly. This offers potential reductions in baggage processing times, which could help avoid flight delays.

The system is being developed by CXR, a research-led organisation set up by airport security specialist Rapiscan Systems. The Manchester mathematicians, funded by CXR and the EPSRC, have been overcoming the mathematical challenges involved in developing the specialised computer hardware and software that underpin the system’s capabilities.

The 3 year project, ‘

X-ray CT
reconstruction algorithms for airport security and process tomography’, is receiving EPSRC funding of nearly £504,000.

Currently, conventional 2D luggage scanning systems may not be able to detect explosives blocked by denser objects. Furthermore, time-consuming manual checks are needed whenever a 2D image is inconclusive.

Although a small number of 3D systems have been introduced, they take three to four times longer to produce an image than a 2D system, because the X-ray source needs to be rotated very slowly around the bag. This leads to a significant interruption in a bag’s journey between check-in desk and aircraft.

Aiming to be as quick and cost-effective as a 2D device and faster than medical 3D scanners, CXR’s new system uses multiple X-ray sources simultaneously. The Manchester team are working on special algorithms that, incorporated into the bespoke software and hardware, enable the separate 2D images produced to be combined into a single 3D image with the required speed and accuracy.