ABANDONED OBJECT DETECTION VIA TEMPORAL CONSISTENCY MODELING AND BACK-TRACING VERIFICATION FOR VISUAL SURVEILLANCE

Abstract: Security of public places is a considerably burning issue. Day by day the issues of mass killing due to bomb explosions are increasing. These bombs are mostly disguised in bags, luggage, etc. The common strategy of sleeper cell is to leave bags or belongings in public area. The security in charge cannot be always vigilant over Camera's footage, hence if an automation is given to Camera itself; that will lessen the chances of such risk. For that the proposed system is processing the video with the image processing using Open CV on .NET platform. If a person is dropping off some bag or any such suspicious thing and leaving it running away, the system will catch this activity and if such bag is immobile for certain time span decided by analyzer, then it will give notification to authority as abandon object using Image processing. Abandoned object detection is an essential requirement in many video surveillance contexts. We introduce an abandoned object detection tool based on a set of possible events and on a set of rules to act upon those events. This implementation is simple and reusable unlike existing techniques. It is implemented using a simple logical reasoning upon textual data, in contrast to image centric processing.

Keywords: Abandoned luggage detection, abandoned object detection, object detection and tracking, video surveillance, left baggage detection, background subtraction.