Astroart Forum

I'm having problems with AstroArt5 over correcting vignetting when applying flats. I thought perhaps it was just the process I was using, but I've had information from another user that they are experiencing the same problem too. Although interestingly they get the problem with one of their scope/camera combos, but they do NOT on another scope/camera setup. They of course use the same processing approach on the sets of data, and one works, and one does not.

Here are a couple of noisy hyper-stretched example images, the first showing the vignetting, and the second showing the overcorrection when flats are applied.
Can anyone offer some advice on what could be causing this, and what we can do to work around the problem?
Cheers

Hello, that is a classical problem that happens when the flat field is not dark-corrected (or at least biasvalue-corrected), the division gives a wrong result:

instead of: corrected image = image / flat
you get: corrected image = image / (flat + offset)
or: (image + offset1) / (image + offset2)

Additional info can be found in the chapter "image calibration".
obviously also the image must be dark corrected. If you don't want to correct the dark frames (recommended for low noise cameras) then at least you must subtract a constant equal to the average value of the dark frame.
P.S.
For non-scientific images "remove gradient" is usually easier than using flat fields.

Thanks for the information, and you are correct that I've not applied any darks here. The multi-step process I use is as follows:

* Make a master bias frame
* Make a master flat, using the master bias frame as the dark frame. I also shift the image by 0.5, 0.5
* Apply the bias corrected flats to the light frames without combining them
* Stack the corrected lights with a sigma combine and a translation+rotation align

So if I'm understanding your post correctly I am bias correcting the flat frames, which should make the flats work? But clearly not in my case.

And yes, the remove gradient function does seem to work well, but it of course wont remove the "dust bunnies" like a flat will.

I suggest to don't use bias frames, they still contain noise and their average value
is different from a correct dark frame. A constant image is better.
To find the problem, process a sample image by hand:

Light, DarkForLight, Flat, DarkForFlat

verify that (Light-DarkForLight) / (Flat-DarkForFlat) corrects the image perfectly.
Then you can remove the darks and replace them with constant images.

I've just done a search and found this old post as I'm having the same problem in AA6. I'm totally lost here. I've been taking my flats, darks and obviously light frames and placing them in the the correct columns ( not as masters though) and I get the same result ......over correction of vignetting . Also I've taken bias frames but can't figure out where to put them. Just re watched the warren Keller DVD but still confused
Could someone please take me through the correct steps.....if they could be so kind

Hello, remember to also use the Darks for Flats, they are mandatory. Biases are useless.
I suggest to make a test by hand, before using the automatic preprocessing:
1) Open one image and correct its dark.
2) Open one flat and correct its own dark.
3) Divide (1) by (2)

I must be really thick or something because I can't get it to work properly

Q1. In the "F.Dark Frame" box do I just put in the same dark frames that I put in the "Dark Frame" box. OR are the F.darks
dark exposures taken at the same length of Flat fields ??
Q2. How do the bias fit in to all of this. Other software I've used is straightforward with boxes to put flats,darks and bias in. I'm totally confused here and really could do with some clarification.

Warren Keller says to put the bias in the dark frame box ......is this along with the darks??

Then there is this over correction of vignetting.......it's really do my head in
I don't know if this makes a difference but it's a OSC ccd I'm using

If anyone can help to clarify this it would be most appreciated

Hi, the dark frames for the flats are different, since they are taken with a much short exposure.
If your flats were taken with an exposure time shorter than 0.5 seconds, then you could use a bias AS a dark frame, since a dark frame of 0.5 seconds is usually identical to a dark frame of 0 seconds, on cooled cameras.

Thanks for that facavi, yes the flats are less than 0.5sec. Just to clarify then, I put the bias in the F Dark Frame box