ADABTS · Automatic Detection of Abnormal Behaviour and Threats in crowded Spaces

ADABTS aims to facilitate the protection of EU citizens, property and infrastructure against threats of terrorism, crime, and riots, by the automatic detection of abnormal human behaviour. Current automatic detection systems have limited functionality, struggling to make inferences about the acceptability of human behaviour. ADABTS aims to address one of the key problems, the definition of abnormal behaviour, by extracting characterizations in realistic security settings based on expert classifications and the analysis of CCTV operator behaviour. A set of behaviour descriptors will be defined and models of threatening behaviour will be built. Algorithms will be developed that detect pre-defined threat behaviours and deviations from normal behaviour. For accurate and robust detection, data from audio and video sensors will be combined with context information. The trade-off between system generality and capability is addressed by extracting a set of intermediate-level behaviour descriptors that are observable from sensor data. This makes the system configurable to specific needs and improves scalability and cost efficiency. A real-time evaluation platform will be developed based on commercial heterogeneous hardware. Such hardware, in rapid development driven by the video game industry, represents a huge potential for high-performance low-cost surveillance systems. The evaluation platform will demonstrate pro-active detection of potential threats. The context will be large-scale events, represented by a football arena, and critical infrastructure, represented by international airports. The consortium partners are well known experts in the area of security and protection of infrastructure. Experts on societal and human aspects are in the consortium. The involvement of stakeholders (security system operators and integrators, police organizations, airports, event organizers) will ensure relevance and exploitation.