Extended model and optimization for FOV of camera network
Bangjun Lei · Jisuanji yingyong yanjiu · 2010
This paper systematically studied how to maximize the FOV(field of view)of an arbitrary multi-camera network,and proposed to gain the global best results by the particle swarm optimization.The FOV improvement was brought by adjusting all cameras’ orientations automatically.Compared with the conventional researches,further considered the LOD(level of details)of each camera and the importance of the targeted scene.Then got a more realistic and more general FOV model.Under this new model,defined a new concept surveillance goodness,and provided the nonlinear equations for the maximization of the goodness.To solve this problem,suggested the particle swarm optimizer and studied it based on the new model for various scenarios.The experiments show that the proposed model and method work much better than the conventional solutions.Both the performance of the algorithm and the scope of applicable scenarios are largely improved.