Emmanuel Bertin

If `gain` is the _effective_ gain (not the original instrumental gain) and `img_data` the pixel value with background subtracted, then an estimate of the standard deviation of the total noise...

No, exposure time is out of the equation because the `GAIN` in the FITS header is assumed to be the _effective_ gain, not the original detector gain before data rescaling....

We are talking about multiple exposures, right? If this is the case, then it depends in what regime. For sources that do not move over more than a few pixels...