Evaluating the accuracy of vehicle tracking data obtained from Unmanned Aerial Vehicles

This paper presents a methodology for tracking moving vehicles that integrates Unmanned Aerial Vehicles with video processing techniques. The authors investigated the usefulness of Unmanned Aerial Vehicles to capture reliable individual vehicle data by using GPS technology as a benchmark. A video processing algorithm for vehicles trajectory acquisition is introduced. The algorithm is based on OpenCV libraries. In order to assess the accuracy of the proposed video processing algorithm an instrumented vehicle was equipped with a high precision GPS. The video capture experiments were performed in two case studies. From the field, about 24,000 positioning data were acquired for the analysis. The results of these experiments highlight the versatility of the Unmanned Aerial Vehicles technology combined with video processing technique in monitoring real traffic data.