Real Time Automatic Object Tracking by Pan-Tilt-Zoom cameras in an IP-Surveillance System
Navneet Kaur · 2012
Pan-tilt-zoom (PTZ) cameras are one of the advanced security cameras in the market. These cameras have the ability to cover a very far field and can acquire high resolution of images. These cameras are deployed mainly for perimeter surveillance applications where the security guards have to monitor the intruders from a long distance. Although there are intrinsic advantages of using pan-tilt-zoom cameras, their application in automatic surveillance systems is still scarce. The difficulty of creating background models for moving cameras and the difficulty of optical geometrical projection models are key reasons for the limited use of pan-tilt-zoom cameras. Geometric calibration is a useful tool to overcome these difficulties. Once developed the background and projection models, it is possible to design system simulators and surveillance methodologies similarly to the ones commonly available for fixed cameras. In this paper, the proposed system can automatically track moving objects. More specific, it is investigated how a system comprised of one pc and a PTZ camera installed within an indoor and outdoor settings can track objects in poor illumination conditions as well. The tracking and calibration results with several image processing techniques in a segmentation framework are combined, through which camera can track the target in real time.