Developing intelligent video surveillance

Ever since terrorist events such as the Madrid train bombing in 2004 and the London underground transport bombing in 2005, Europe has been forced to be more vigilant about its infrastructure. Current closed-circuit television (CCTV) cameras can only go so far in detecting suspect or abnormal behaviour, implying a need for more sophisticated technology that would pre-empt such apocalyptic scenarios.

The EU-funded project SAMURAI (Suspicious and Abnormal Behaviour Monitoring Using a Network of Cameras & Sensors for Situation Awareness Enhancement) set out to develop a smarter and more advanced monitoring system that detects suspicious behaviour in real time by monitoring a vast network of cameras. The new technology is based not only on CCTV cameras but also on the use of various sensors, combining several streams of data together for a more comprehensive picture.

Current security CCTV systems process video using pre-established rules that yield a high number of false alarms. SAMURAI devised a system that determines abnormal and acceptable behaviour based on feedback from system operators.

The innovative surveillance system tracks objects and people through multiple camera views and lighting levels, relying on shapes and movement patterns to identify people. It works with a variety of sensors such as mobile wearable audio and video recorders that are fitted onto security teams to improve surveillance.

The technology focuses on real-time images and provides continuous context-based information to security personnel. System operators can also instruct the system using feedback so that it does not generate false alarms.

SAMURAI enhanced the capabilities of CCTV, enabling intelligent, adaptive and rapid security surveillance. People will ultimately feel safer and more secure in public spaces.