Tracking of Moving Vehicles with a UAV
Min-Hyuck Lee, Seokwon Yeom · 2018
Multiple object tracking is important for video surveillance. This paper addresses the tracking of moving vehicles with an unmanned aerial vehicle (UAV). Moving cars are segmented to generate the centroids of objects. The centroids become inputs for tracking. Tracking is performed with Kalman filtering which estimates the dynamic states of targets. In the experiment, a UAV generate video sequences of three cars driving on roads. Experimental results show that the proposed method well track the moving cars.