Real Time Tracking of Moving Pedestrians
Juan Li, Chunfu Shao, Wangtu Xu, Hao Yue · 2009
Pedestrian detection is an important part of intelligent transportation systems (ITS) application. Many researches have been contributed to apply image processing technologies to pedestrian detection. A new robust method for pedestrian tracking is proposed in this paper. Pedestrian tracking is achieved by using feature fusion and prediction methodology. The proposed method integrates spatial position, shape and color information to track pedestrians. The trajectories obtained from camera are incorporated by the Kalman filter to determine the search area. The presented tracking method is tested under real traffic scenarios. Elaborate experiment results show that integrating simple features makes the tracking effective and the Kalman filter improves the tracking accuracy and efficiency.