Robust multiple human tracking using particle swarm optimization and the Kalman filter on full occlusion conditions

Visual surveillance in crowded scenes, especially for humans, has recently been one of the most active research topics in machine vision because of its applications such as deter and response to crime, suspicious activities, terrorism or human behavior recognition. One of the most important problems in multiple human tracking is the occlusion problem. When the number of humans has an occlusion with each other or the background, the tracker should track them correctly. In this paper, we use particle swarm optimization (PSO) as a tracker, in addition to the Kalman filter and some other mathematical equations to solve the occlusion problem which the occlusion can be partially or completely. Experimental results on several real videos sequences from different conditions have shown the effectiveness of our approach.