MEAN SHIFT AND TRACKING ALGORITHMS IN VIDEO STREAMS

Recognition of hand gesture is very important to communicate in sign language. In this paper, is presented a modified method for object tracking and hand gesturerecognition. There are two common algorithms for object tracking: the mean-shift and the motion detection algorithm. In this approach, the advantages of the two algorithms are used toimprove the hand tracking. At first, the motion detection algorithm subtracts the previous frame from the current frameto obtain the changes between two images and white pixels (motion level) are detected by using the threshold level. Themean shift algorithm is only utilized for tracking the hand motion. If each one of above algorithm is applied, there are several problems for hand tracking in some parts. Therefore, ithas used the combination of two algorithms to compensate the weaknesses of mentioned algorithms. Simulation results show the proposed method can easily track the hand motion

کلیدواژه ها:

Hand gesture recognition ، Hand tracking ، Mean shift ، Motion detection ، sign language