摘要: Video-based crowd behaviour detection aims at tackling challenging problems such as automating and identifying changing behaviours under complex real life situations. In this paper, real-time anomaly algorithms have been investigated. Based on the spatio-temporal video volume concept, an innovative texture model has proposed in research for its rich pattern characteristics. Through extracting integrating those textures from surveillance recordings, a redundancy wavelet transformation-based feature space can be deployed behavioural template matching. Experiment shows that abnormality appearing scenes identified fashion by devised method. This new approach is envisaged to facilitate wide spectrum of analysis applications through current Closed-Circuit Television (CCTV)-based systems.