GEDetector: Early Detection of Gathering Events Based on Cluster Containment Join in Trajectory Streams

Existing trajectory patterns, such as flock, convoy, swarm, and gathering, are to detect moving clusters staying or travelling together for a certain time period. But these patterns model group movement behaviors after moving objects’ gathering together. This may result in loosing golden opportunities to detect emergency incidents earlier, such as traffic congestion and serious stampedes. In this work, we propose a novel group pattern, called converging, which can model converging behaviors of moving objects. As a proof-of-concept, we implemented a visual analytic system GEDetector based on trajectory streams to detect gathering events as early as possible. A user-friendly interface is designed to help users gain insights into gathering events from spatial and temporal aspects. Finally, we demonstrate the effectiveness and efficiency of our system by using a real world dataset.