Large-scale surveillance system: detection and tracking of suspicious motion patterns in crowded traffic scenes

The worldwide increasing sentiment of insecurity gave birth to a new era, shaking thereby the intelligent videosurveillance systems design and deployment. The large-scale use of these means has prompted the creation of new needs in terms of analysis and interpretation. For this purpose, behavior recognition and scene understanding related applications have become more captivating to a significant number of computer vision researchers, particularly when crowded scenes are concerned. So far, motion analysis and tracking remain challenging due to significant visual ambiguities, which encourage looking into further keys. By this work, we present a new framework to recognize various motion patterns, extract abnormal behaviors and track them over a multi-camera traffic surveillance system. We apply a density-based technique to cluster motion vectors produced by optical flow, and compare them with motion pattern models defined earlier. Non-identified clusters are treated as suspicious and simultaneously tracked over an overlapping camera network for as long as possible. To aiming the network configuration, we designed an active camera scheduling strategy where camera assignment was realized via an improved Weighted Round-Robin algorithm. To validate our approach, experiment results are presented and discussed.