Indoor security system design and implementation using depth information

A real time security system is used in residential, commercial, industrial, and military properties for protection against theft as well as personal protection against intruders. In this paper a real time indoor security system is designed which is capable of extracting any object that passes through its range and provide a security alarm when the extracted object enters the restricted area. For the extraction of the object a well known background subtraction process is used. The technique Stereoscopy is used for recording and representing stereoscopic images with the help of stereo camera. Stereo images are used to estimate the distance of the object with the help of a proposed mathematical model between the disparity image and the object distance. A threshold value is previously defined and an alarm is blown when the object exceed the minimum permissible distance. The proposed system implementation with OpenCV makes it well suited for real-time application.