Visual Traffic Load Sensor for Emission Estimation

Traffic enumeration is a frequently used approach in smart city applications for estimating the traffic load of the road network. Many solutions use a camera and state-of-the-art image processing technologies. As image processing is relative resource consuming, either very simple methods are used, or the processing is performed off-site, on a central computer. The approach presented in this paper employs a powerful embedded system to process the video images in real-time, so that the communication traffic is limited to the transmission of retrieved statistics. In order to fit into the limited resources and still allow vehicle size and speed measurements, special data representations and processing pipeline has to be used. This paper presents the operation principles of the sensor, followed by the corresponding evaluation results.