Title: Optimisation collaborative sense of event area in smart camera networks

Authors: Jun Liu; Xu Lu; Jianhua Wang

Addresses: School of Automation, Guangdong Polytechnic Normal University, Tianhe, Guangzhou 510665, China ' School of Automation, Guangdong Polytechnic Normal University, Tianhe, Guangzhou 510665, China ' School of Electrical and Electronic Engineering, South China Agricultural University, Tianhe, Guangzhou 510642, China

Abstract: Smart camera networks have good application prospects in the future. Smart cameras as video sensors can adjust the focal length and 360° rotations so that they can flexibly observe the surrounding event area. It is an interesting issue how to schedule multiple video sensors for high-quality monitoring of multiple event points in the monitoring area; it is the same as coverage optimisation problem. In fact, this is a combinational optimisation problem. The optimal solution is achieved by adjusting the parameters of multiple sensing nodes. In this paper, the mathematical density condition of sense area covered is deduced firstly. Then, the smart sensor coverage optimisation problem is transformed into an NP-hard combinatorial optimisation issue. A heuristic greedy algorithm and a centralised hybrid algorithm combining the artificial bee colony algorithm with different algorithms are proposed. The algorithms are capable of solving this optimisation problem in different scenes. Finally, some simulations are conducted to contrast the performance of different algorithms. Experimental results show that the algorithms performed better than other similar algorithms in terms of time convergence and sense accuracy with different scenarios.

Keywords: VSNs; visual sensor networks; pan-tilt-zoom; priority event; artificial bee colony; heuristic algorithm.

DOI: 10.1504/IJCAT.2019.099501

International Journal of Computer Applications in Technology, 2019 Vol.60 No.1, pp.1 - 11

Received: 13 Jun 2018
Accepted: 09 Jul 2018
Published online: 07 May 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article