Traffic Data Collection under Mixed Traffic Conditions Using Video Image Processing

Abstract

Traffic data collection under mixed traffic conditions is one of the major problems faced by researchers as well as traffic regulatory authorities. Study and analysis of traffic behavior is critically dependent on the availability of observed traffic data. For mixed traffic observed in developing countries, no suitable tool is available for this purpose. Keeping in view these necessities and problems in data collection, a novel offline image processing-based data collection system, suitable for mixed traffic conditions, is developed. Its underlying ability to detect, track, and classify vehicles makes it useful in collecting traffic data under varying traffic conditions. This system can automatically analyze traffic videos and provide macroscopic traffic characteristics such as classified vehicle flows, average vehicle speeds and average occupancies, and microscopic characteristics such as individual vehicle trajectories, lateral, and longitudinal spacing. It is observed that this new system is working well even under congested mixed traffic conditions.