Person Scanner Detects Hidden Weapons
Germany’s weapons laws are set to become stricter, especially concerning knives. However, enforcing these regulations presents challenges related to manpower. A security scanner developed by Fraunhofer FHR allows for the detection of hidden weapons in crowds – essentially while people walk by.
The knife attack in Solingen has shaken Germany – the federal government is therefore planning to tighten gun laws. Among other things, knives are to be banned at public festivals, sports events, and on buses and trains. However, monitoring this will require a lot of staff. A security scanner for civilian use from Fraunhofer FHR comes at just the right time: unlike standard scanners at airports, this model can check a high throughput of people, such as those on their way to the subway. Stopping and waiting in line is unnecessary; people are checked while walking by.
MIMO radar reveals concealed items worn underneath clothing.
The scanners are set up on either side of a passageway, such as at the entrance of an escalator or stairs. As a person walks through, they are automatically scanned. The core of the scanner is a MIMO radar system, short for "multiple input, multiple output": it can generate an image of the person using its 352 transmitting antennas and 528 receiving antennas. "Since radar waves are used in the frequency range of 6 to 10.5 gigahertz, they can easily detect items hidden under clothing," explains Patrick Wallrath, group leader of 3D signal processing at Fraunhofer FHR. The principle is that radar waves are reflected off the person and, if present, the knife or other weapons, and received by the antennas. Each of these measurements provides 92,864 signals, from which an image is reconstructed. To do this, Fraunhofer SCAI implemented the algorithms of Fraunhofer FHR on a graphics card. "Since it takes some time to generate an image while the person continues to move, our system compensates for the person's speed," says Wallrath. The scanner produces a total of five images per second, and the evaluation is done in real-time, handled by a French partner using artificial intelligence.
Prototype tested at subway station in Rome
The prototype of the scanner is already complete: it was successfully tested for two days at a subway station in Rome. The system was developed in the NATOSPS (Science for Peace and Security) project "DEXTER" as a component of a larger system made up of three subsystems: Along with the person scanner MIC from Fraunhofer FHR, there is a technology that can detect traces of explosives on surfaces like hands and luggage, as well as a system that can track suspicious individuals in crowds until security personnel can intervene. The goal of the project was to conduct security checks at critical infrastructures such as subway stations.