Traffic Flow Monitoring in Crowded Cities.
John A. Quinn, Rose Nakibuule · 2010
Traffic monitoring systems usually make assumptions about the movement of vehicles, such as that they drive in dedicated lanes, and that those lanes rarely include non-vehicle clutter. Urban settings within developing countries often present extremely chaotic traffic scenar-ios which make these assumptions unrealistic. We show how a standard approach to traffic monitoring can be made more robust by using probabilistic inference, and in such a way that we bypass the need for vehicle seg-mentation. Instead of tracking individual vehicles but treat a lane of traffic as a fluid and estimate the rate of flow. Our modelling of uncertainty allows us to accu-rately monitor traffic flow even in the presence of sub-stantial clutter.