Color particle filter based object tracking using frame segmentation in CIELab* and HSV color spaces

Object tracking is a significant process used in many application areas. This paper presents a color based particle filter algorithm used for object tracking in video sequences for CIELab and HSV color space. Initial state of particles represents the position of pixels in the object to be tracked. The initial state is updated and weights are assigned to the particles depending on the value of the likelihood distance. Likelihood is computed based on color similarity with target region. The process leads to particle degeneracy and it is handled by using resampling stage. The method provides reliable and effective tracking results for many indoor and outdoor video sequences.