Abandoned Object Detection via Visual Surveillance

Abstract: In this paper we are processing the live feed of the CCTV camera with the image processing. If a person is dropping off some bag or any such suspicious thing and leaving it running away, the camera will catch this activity and if such bag is untouched for some time span decided by analyzer then it will give notification to authority. Minimum the time span probability stays 50-50, but as time span increases the probability of that thing being abandon or hazardous increases. We introduce a framework to identify static foreground regions based on the transition of code patterns and to verify whether the candidate regions content abandoned objects by analyzing the back-traced trajectories of luggage owners. Hence the incident can be avoided in that case.

Keywords: Abandoned luggage detection, abandoned object detection, object detection and tracking.