Hi folks, I've seen papers where the SNR for a scene has been estimated for a scene using the data itself. These papers have brief discussions of how they did it, but it didn't quite make sense to me. Can anyone give a fairly complete outline of how this is done? Jeff