Complex Event Detection in Video and Communications

This Small Business Technology Transfer (STTR) Phase-I project will constitute design of algorithms, toolsets, and interfaces to detect and visualize complex events from video and communications data. Our proposed system would be based on a Service Oriented Architecture (SOA) which will allow users to integrate services and data from multiple sources and operate in a networked environment. The key innovations in the proposed effort are: 1) automated detection of rule-based and unusual events that may span multiple sensors, 2) detection of network anomaly and intrusion detection from communications data, 3) reliable complex event and anomaly detection using data from multiple cameras and network services using a detection oriented Complex Event Processing (CEP) engine that allows fusion of data from various SOA-based client services, 4) enhanced user interface based on NASA's WorldWind geo-browser plug-in for wide area situational awareness of the operator. In addition, metrics for complex activity detection will be investigated and a thorough performance analysis of the proposed innovation will be carried out to ensure robustness. This will ensure that a scalable and robust baseline complex event detection system is developed by the end of Phase I.