Background subtraction-based multiple object tracking using particle filter

In surveillance system, many vision based methods are utilized for multiple objects. Some of the methods used for tracking include background modeling, Kalman filter, particle filter etc. The proposed method combines both the background modeling and particle filter to track multiple objects. This method consists of two steps: background subtraction and application of particle filter. Background subtraction restricts the area of search so that the particle filter performs efficiently. Different colored particles are displayed on the object for tracking. After applying the proposed method on a video having multiple objects, successful results were obtained.