Abstract
Traffic flow surveillance for traffic control and transportation planning requires fixed-point continuous observation of exact vehicle movements. The high altitude platforms have a great potential with vehicle maneuvers observation. In this article, we develop a vehicle maneuvers recognition technique with sequential images from high altitude platforms. We exploit spatio-temporal clustering method. Employing the two features, that are background subtraction value and optical flow, the pixels that are adjacent and have similar features are grouped. Through some applications to real sequential images from a helicopter, the effectiveness of the proposed method is confirmed.
