purpose: Creates a master screen flat from a set of screenflat images.
description: After bias subtraction, the input frames are averaged. The overscan regions, if present, are removed from the result. The combined frame is then normalised dividing it by its large scale illumination trend. The large scale trend is obtained by applying a median filter with a large kernel. The master frame has 2 extensions, with the actual product being in the first extension and its errors being in the second extension.
At the end, the new master frame is compared with the corresponding certified master frame from the calibration database.