A Multi-Camera Surveillance System that Estimates Quality-of-View Measurement

In this paper, we propose a multi-camera video surveillance system with automatic camera selection. A new confidence measure, quality-of-view (QOV), is defined to automatically evaluate the camera´s view performance for each time instant. This measure takes into account view angle and distance from subjects. By comparing each camera´s QOVs, the system can select the most appropriate cameras to perform specific tasks. We also present an approach to determine the minimum number of cameras and their layout in a convex polygonal room under specific QOV constraints. Finally, we implement an experimental surveillance system, to confirm the stability of our algorithm and validate the critical underlying concepts of QOV.