Optimizing Visual Sensor Coverage Overlaps for Multiview Surveillance Systems

Modern surveillance systems rely on coverage overlapping to attain multi-viewing capabilities and to support coverage redundancy. This paper examines the coverage overlapping configurations in visual surveillance systems. This paper proposes a robust dynamic programming framework to optimize visual surveillance sensor coverage overlaps. Based on visual sensor parameters information, and the features of the area to be monitored, this paper uses a deterministic modeling approach to model the sensor coverage in a 2-D space. Then, the minimization and the maximization arrangements of the coverage overlapping are formulated as discrete optimization problems. The obtained solutions from the dynamic programming technique are evaluated with respect to local and global greedy search algorithms. The results reveal the feasibility of the proposed technique compared with the benchmarked optimization methods in terms of the amount of coverage redundancy.