Vision based indoor positioning for intelligent buildings

With the surveillance camera become more and more popular, people need to observe a large number of monitor at one time. In this paper, we propose a vision based indoor positioning for intelligent buildings to solve this problem. We capture a frame and make the Gaussian mixture learning for video background model. Then we do the background subtraction and get the foreground object. After we have the foreground object, we track objects and use the direct linear transform to integrate camera information to a bird's-eye map. finally all the moving objects are traced and displayed on the bird's-eye view.