Abstract:
Motion estimation is an important subject in the field of computer vision. Various techniques, formulated for motion detection are used by many systems for different purposes like warning, alarm systems, video compression, speed measurement and so on. This thesis report unfolds all the details needed to understand various aspects of a moving object in a sequence of video frames. Background Learning module is the first and the foremost important module, in terms of its functionality. Background is the key frame, which includes every thing in the environment under consideration, except the moving objects. Efficiency of the background-learning module has its effect on the overall system of motion detection, as every thing else is determined accordingly using the background image. To add to the good work done by various organizations in this field, we have come up with our proposed algorithms that promise to take the research in to a broader perspective.
