A quality-enhancing coverage scheme for camera sensor networks

Exploiting camera sensors to conduct intruder detection has attracted a lot of research attention. Different from the traditional sensor with omni-directional sensing model, a camera sensor usually has a specified direction with a fixed sensing angle of sensing area. The sensing quality of coverage is critical to the application of coverage scheme. In this paper, we first investigate the complete coverage issue with two alternative layouts of sensing area by 2 camera sensors. Considering the weighted image quality and the importance of sensing area, we propose a quality-enhancing coverage scheme for camera sensor networks, QCC, to improve the coverage performance. We then mathematically present a geometrical probability-based analysis to theoretically evaluate the performance of intruder detection approach. Furthermore, the extension of QCC scheme, QCC-D, is proposed to cover the sensing areas with differentiated importance. Through extensive simulations, our scheme is demonstrated to outperform the best known coverage algorithms, in terms of both overlap area ratio and total weighted quality.