Abstract. Unmanned Aerial Vehicles (UAVs) can be used for several applications such as: precision agriculture, environmental monitoring and management, smart cities, planning, and others. Specifically for smart cities, a very desirable application is the identification and tracking of vehicles in roads for planning and traffic optimization. In this paper, a computational approach is proposed to address this problem. Our solution has the proposed pipeline: (a) background subtraction is applied to extract the vehicles in motion; (b) tracking method is used to keep a history of trajectories for each vehicle. The preliminary results were obtained from an UAV prototype, showing promising results for traffic counting and flow analysis.