S. Fleck, C. Vollrath, F. Walter, and W. Sraβer (Germany)
smart camera, particle filter, video surveillance, camera network, visual tracking, surveillance system
Surveillance systems based on distributed sensor networks are massively emerging today, as the interest in enhanced safety in this ever changing world gets more actual than ever. Traditional “CCTV” surveillance systems with their centralized processing (compression) and recording archi tecture together with a simple multi-monitor visualization of the raw video streams bear several drawbacks and lim itations. The necessary communication bandwidth to each camera and the computational requirements on the cen tralized servers strongly limit such systems in terms of the expandability, installation size (cable lengths) and spa tial/temporal resolution of each camera. Additionally, the visualization is counter intuitive and fatiguing due to the massive load of raw video data. We present a distributed network of smart cameras and its integrated visualization. The smart cameras’ track ing results are embedded in a common 3D environment as live textures and can be viewed from arbitrary perspec tives. Also a georeferenced live visualization embedded in Google Earth is presented. This offers a visualization more abstract from the camera perspective yet more intuitive in terms of integration.
Important Links:
Go Back
