Autocorrelation of gradients based violence detection in surveillance videos

Automated monitoring of videos is becoming mandatory due to its widespread applications over public and private domains. Especially, research over detecting anomalous human behavior in crowded scenes has created much attention among computer vision researchers. Understanding patterns in crowded scenes is always challenging due to the rapid movement of the crowd, occlusions and cluttered backgrounds. In this work, we explore spatio-temporal autocorrelation of gradient-based features to efficiently recognize violent activities in crowded scenes. A discriminative classifier is then used to recognize violent actions in videos. Experimental results have shown improved performance of the proposed approach when compared to existing state-of-art-approaches. (C) 2020 The Korean Institute of Communications and Information Sciences (KICS). Publishing services by Elsevier B.V.