In most current surveillance systems, the raw live video stream from a large number of cameras is displayed on a set of monitors, and security personnel respond to situations as they see them. Activity of interest must be manually tracked, and when an individual leaves one camera's view observers must switch cameras and viewpoints. To improve upon this approach, Sven Fleck and his colleagues at the University of Tübingen (Tübingen, Germany; www.gris.uni-tuebingen.de) have developed what they say is an intuitive visualization approach in which the results of a surveillance system-the person's path-are reflected in one consistent and georeferenced 3-D world model that is ubiquitously accessible.
On display at VISION 2006 in Stuttgart, Germany, the SmartSurv surveillance system is composed of a distributed network of mvBlueCougar intelligent cameras from Matrix Vision (Oppenweiler, Germany; www.matrix-vision.de) that are capable of tracking objects in real time. The visualization capability is provided by a mobile platform containing a laser scanner and panoramic camera or within Google Earth. As the entire tracking is embedded inside each smart camera node, only limited bandwidth is necessary, which makes the use of Ethernet possible. The potential number of cameras could reach into the thousands. Fleck says the group is looking at many applications for which the system could be commercialized, including tracking shopping patterns and monitoring the elderly.
