Design of quad-rotor target tracking system

Abstract

The accuracy of tracking based on Camshift would decrease due to the similarity between target color and background color or the target is obscured. For the above problems, improved target tracking algorithm based on Camshift is proposed in this paper. The Camshift algorithm is improved by using the contour features of the target, and Camshift search window is updated according to the contour feature of the target. Thus, interference of background and strong light is weakened. Kalman filtering algorithm is used to predict the motion state of the tracking target, enhancing the efficiency of tracking when the tracking target is obscured. Experiments show that Camshift is combined with the contour feature of target and make the tracking more effectively under the conditions of background. And the Kalman filtering algorithm is used to predict position of the target to make the tracking effectively when the target is obscured.