Traffic Anomalous Region Detection Model
Yu-Ling Hsueh, Wei-Li Lai, Chia-Chun Lin, Pierre Pascal Lindenberg · 2016
In this paper, we propose a traffic anomalous region detection system based on a heat diffusion model to detect anomalous regions. We use the heat diffusion model to infer the origin of a traffic anomaly by observing the level of traffic flow spread along the road networks in a similar way to the heat energy transferring from a high temperature region to a lower one. Furthermore, we discuss the relations of the neighboring regions to determine the candidates for events occurring in these regions. The weights used in the model represent the environmental conditions which affect the “conductibility” of the traffic flow. We implement our method and compare it with an existing work. The experimental results show that our estimation of the traffic flow distribution is much closer to the actual sensor records than that of the existing work.