how to detect new shape use background subtractor
hello now background subtraction work on video learns from all image the background and if new shape come it is foreground , i want to detect shape , i know the shape located in the center of image +- some pixels, the backrgound is in the edges of the image +- pixels how i can use background subtraction for find the shape i want to use Gaussian mixture model such as in http://mateuszstankiewicz.eu/?p=189#comment-16981
http://opencv-code.com/tutorials/detecting-simple-shapes-in-an-image/
there's a sample here , also bookmark the docs
since all you get from the bg-segmentation is a binary mask, next step would be to isolate the contour ( see balaji's link above)