Unattended and stolen object detection based on relocating of existing object

Abstract

© 2014 IEEE. This paper proposes a new design and implementation method in supporting a smart surveillance system that can automatically detect abandoned and stolen objects in public places such as bus stations, train stations or airports. The developing system is implemented by using image processing techniques. In the circumstance such as when suspicious events (i.e. left unattended or stolen objects) have been detected, the system will alert to people responsible for the role such as security guards or security staff. The detection process consists of four major components: 1) Video Acquisition 2) Video Processing 3) Event Detection and 4) Result Presentation. The experiment will be conducted in order to access the following qualities: 1) Usability: to verify that the system can detect the object and recognize the events, and 2) Correctness: to measure the accuracy of the system. Finally, the data sets are tested in the experimental result can measure and represent the correctness of system by percentage. The correctness of object classification is approximately 76%, and the correctness of event classification 83%.