if I give input as a image or video then my algorithm or code should identify the circular objects in that image or video. For example, in provided image my algorithm or code should identify the ball. I tried Hough circle transform to detect them but it failed to detect it