Abnormal Behavior Detection Based on the Potential Field Forces in Surveillance Scenarios
Yong Qiang, Shumin Fei, Liuwen Li · Journal of Physics Conference Series · 2020
Abstract In a social environment, there is a natural mutual force between people. This paper proposes a method for detecting group abnormal behavior based on the potential field method. The movement of a pedestrian is described as the movement towards the target under the combined action of several forces. Forces on pedestrians include the attractive force for themselves and the repulsive force between pedestrians. By calculating the resultant force between them, the behavior of pedestrians is judged according to a set threshold. It is verified through experiments on public data sets that our method has high accuracy and robustness.