Particle Swarm Optimization Inspired Probability Algorithm for Optimal Camera Network Placement

Abstract

In this paper, a novel method based on binary Particle Swarm Optimization (BPSO) inspired probability is proposed to solve the camera network placement problem. Ensuring accurate visual coverage of the monitoring space with a minimum number of cameras is sought. The visual coverage is defined by realistic and consistent assumptions taking into account camera characteristics. In total, nine evolutionary-like algorithms based on BPSO, Simulated Annealing (SA), Tabu Search (TS) and genetic techniques are adapted to solve this visual coverage based camera network placement problem. All these techniques are introduced in a new and effective framework dealing with constrained optimizations. The performance of BPSO inspired probability technique is compared with the performances of the stochastic variants (e.g., genetic algorithms-based or immune systems-based) of optimization based particle swarm algorithms. Simulation results for 2-D and 3-D scenarios show the efficiency of the proposed technique. Indeed, for a large-scale dimension case, BPSO inspired probability gives better results than the ones obtained by adapting all other variants of BPSO, SA, TS, and genetic techniques.