Missing Object Detection using image processing

Abstract

CCTV cameras are used worldwide to monitor various important public and private places like banks, railway stations, museums. Usually the operators who are in the camera room need to be watchful to monitor several camera videos that are fed from various locations. Therefore the manual monitoring of camera scenes increases the chance of omitting some important incident (an incident that requires quick actions) that is occurred but not being noticed by anyone. Since this manual monitoring of cameras hinder the capabilities of cameras, because even though the camera captured the incident the human operator might not notice whether something happened somewhere in the scene under the surveillance. It is very important and easy if the camera is intelligent and can detect actions on its own and alert the operator, in other words this can be thought as a way of extending the camera capabilities to provide a better service. This project uses the camera as an eye and tries to continuously watch selected objects in the scene .In case some object is moved away from its original location or some object is completely missing that will be alerted to the user or any other external system (alarming system). When the complete system is considered, the system will be capable of watching a given objects and identify its displacement and in case it find that particular object is missing the camera will alert the relevant system/person about that incident. The system consists of image capturing module which captures frames from a video sequence and a motion recorder module which will record the motions for the particular object (the desired object that the system is supposed to keep its eye on).Alert the state of the missing object incase the object is moved out from the scene on which the camera is focusing. The capturing module input frames into the motion recorder module which will watch the target object continuously simultaneously the position of the object will be validated against the initial position of the object in order to calculate the displacement. The core concept of the system will be to continuously watch a given target object to verify whether its being displaced and/or missing with respect to its initial location in the scene and then alert in the case it finds the object is missing .