Abnormal activity detection in video sequences using learnt probability densities - TENCON 2003. Conference on Convergent Technologies for Asia-Pacific Region

Absfvact-Video surveillance is concemed with identifying 2. FEATURE XTRACTION AND PROCESSING abnormal or unusual activity at a scene. In this paper, we develop stochastic models to characterize the normal activities in a scene. Given video sequences of normal activity, probabilistic models are leamt to describe the normal motion in the scene. For any new video sequences motion trajectories are extracted and evaluated using these learnt probabilistic models to identify if they are abnormal or not. In this paper, we have employed the commonly used prototype based representation to describe the movement of individual objects. The model parameters are estimated in the Maximum-Likelihood framework. There are several static features that are useful in identifying abnormal activity. However, in a generic video surveillance system, information about the the strongest clue to detecting abnormal activities. Many surveillance algorithms use motion information as the main feature [4], [ 5 ] . Our algorithms ,are aihed at'detecting motion patterns ofobjects that are abnormal in a given scene.