Automated image tracking based on the CAMSHIFT algorithm with adaboost and target trajectory and size estimation
Chi-Juang Hsieh, Kai‐Yew Lum · 2014
In order to overcome a shortcoming of traditional CAMSHIFT which requires manual target designation and object color similarity between frames, this paper proposes an image tracking algorithm by a combination of CAMSHIFT with Adaboost object detection and the Kalman estimator. The proposed method alleviates loss-of-track problems caused by accelerating target motion and color noise. Robustness against occlusion is also improved, while the number of iterations that CAMSHIFT requires is reduced.