Hi,
Let's say I want to detect the (x,y) coordinates of some number (>1) of moving bright points from a camera image using simple thresholding. Also assume that I can obtain a ROI around the points of interest.
As they move further away, their size reduces and hence makes them harder to detect.
So my question is:
Is is possible to zoom into the ROI ( by somehow calculating the zoom factor ) , detect the (x,y) points of the center of bright points by using something like image moments, and then reverse map the (x,y) points to (x,y) in the original image (the one without any zoom) ?