Probability-based optimal coverage of PTZ camera networks

Distributed camera networks represent an emerging trend for the incorporation of pan-tilt-zoom (PTZ) cameras in next generation large-scale wireless surveillance systems. The adoption of PTZ cameras with dynamic field of view nature can increase the resolution of mobile targets and adapt the network coverage significantly. However, leverage these benefits calls for more configuration cost and sophisticated solutions. In this paper, we consider the specific problem of coverage problem in terms of configuration cost and coverage quality in PTZ camera networks. We argue that the coverage algorithms can exploit the dynamic field of view nature of PTZ cameras to reduce the configuration cost by taking into account the heterogeneous coverage quality demands for different targets in a surveillance application. We propose a new coverage definition, termed probability coverage (p-coverage), according to the patrolling property of PTZ cameras. A optimal coverage issues, min-quantity, is formulated underlying the proposed p-coverage framework. A greedy-based heuristic algorithm is proposed to solve the above probability-based optimal coverage problem. The proposed solutions are evaluated to highlight the relative significance of configuration cost gain with respect to diverse coverage demand.