Biologically inspired object tracking: A modular approach with distributed particle like sensors

摘要: Innovation in computing technology especially embedded processors capabilities has opened up the boundaries for an enormous potential to facilitate automation domestic consumer market normally sake of added safety and reliability [1][2]. These applications also demand real time processing example as required Surveillance. Although majority research computer vision revolves around adopting analytical approach this problem, but trackers solely dependent on such algorithms e.g. Kalman filters [3] are not always feasible due presence non-linearity or non gaussian like state space distributions. On other hand particles based [4] requires intensive hence could fail comply with situations needing interventions. This paper is attempt highlight issues propose a modular evolutionary tracking algorithm particle swarm optimization [5], distributed sensors relevant search that dynamically changes complexity scene conditions. To reduce amount global best first formulated using selective histogram back projection algorithm. The prediction step projects prior shape contour object interest onto current frame by initial positioning forming explores Fig1. Therefore controlled dynamics inspired explore only dramatically decreases no our experiments compared traditional proves more robust thus facilitates video.