Toward mining anomalous behavior from big moving trajectories in surveillance video

With the dramatic growth of using video cameras for applications of public surveillances in recent years, detection of public threats or security issues on surveillances becomes possible nowadays. How to identify anomalous behavior from surveillance videos has been identified as an effective manner for detecting critical events in the public avenue. We in this paper discuss a new application paradigm to identify anomalous moving behavior by utilizing techniques of mining trajectories which are extracted from moving objects in the surveillance video. Our experimental results show the effectiveness of our proposed algorithms, demonstrating its promising applicability in the big data era.