Human limb motion real-time tracking based on CamShift for intelligent rehabilitation system

Intelligent rehabilitation system has become a hot topic. In home rehabilitation system, the real-time tracking of human motion is very important. In this paper, it presents a modified CamShift (continuously adaptive mean shift) algorithm, which can track human limb motion in real time. Considering the limb motion and selected color probability distribution all together, we put forwards the modified algorithm, which enhances anti-noise ability and avoids distracters. Finally, based on the OpenCV computer vision library, use CamShift tracking algorithm to achieve limb motion tracking. Experiment results show this algorithm is fast, accurate, and robust. It can be used in other region tracking special color.