Motion Object Detection System for Video Surveillance using Block Based Background Subtraction Method

Closed-circuit television (CCTV) is the use of video cameras to transmit a signal to a specific area, on a restricted set of monitors. This technology is being used for supervision in areas such as banks, airports, hospitals, and shopping malls etc. The installation of the CCTV helps to avoid crime and may assist in the solution of cases, influenced by the need for increased security. Smart video surveillance systems are proficient of enhancing situational awareness up to great amount. In planned system we have stated, Smart CCTV technology, which reviews the circumstances and alert the administrator directly and respond immediately. This project makes use of OpenCV library to capture camera images and detect intrusion using image comparison technique (block based background subtraction method). In this technique, a captured image has been divided into the standard sized blocks and then, only the required block of the earlier frame image is restructured in real-time. Once the comparison is completed and an interruption is found, it sends the video from server to distant administrator over android phone. Administrator can then take suitable action and feasible safety measures. We have designed the detect ability for a moving object region by rapidly creating a background image even if it is difficult to click a background image or we do not have the template image prepared in advance. It is also superior for observation of many places at the same time with only a single CCTV system since it is particularly suitable for sudden scene changes.