Multiple Target Tracking With the Help of Mean Shift Algorithm

The multi moving targets tracking are a typical job in the field of visual surveillance. The main difficulties in targets tracking are fast motion of the target, suddenly velocity variations, clutters, complex object shapes and occlusions, etc. Target tracking are basically explain in to three main parts: point tracking, kernel tracking and silhouette tracking. Here, the authors can track multiple targets with the help of mean shift algorithm. For tracking purpose, the mean shift method requires a kernel that mean's a circle or a rectangle or an ellipse region.