MOTION DETECTION IN A VIDEO SEQUENCE USING BACKGROUND SUBTRACTION BY 3D STATISTICAL MODELING IN THE VISIBLE SPECTRUM: HOME SURVEILLANCE CAMERA

Motion recognition is one of the key applications of motion detection, which involves image processing. For this reason, the proposed work is merely an extension of the final year project. It will involve a detailed study of a method for automatic, real-time recognition of motion and identity in a video stream. Recognizing the suspicious movement of an object or person in an image is a major challenge in the field of camera-based surveillance. This recognition of suspicious object movement involves several phases: detection, tracking, creation of a database or vocabulary against which movements can be compared in order to be recognized and qualified as suspicious or not. In this work, we focus on the phase of detecting a moving object or person in a video sequence. We study a number of existing methods for detecting the movement of objects or people in a video sequence, in order to select the best among them for possible implementation.