Integrating different subs

[niccoc1603]

I need advice on the following: I have two sets of subs taken on two nights, they have slightly different gains and exposure lenghts (200vs240s) I want to integrate all the subs together, there is no need for HDR here, should I just calibrate each set independently and then just integrate all the subs together?

[Niall Saunders]

Why are you using different settings for GAIN? This should be a number that, along with OFFSET, should be established emprically for your imager, tested, and then it should be left alone unless you find that new results of the empirical tests suggest that the new data is better than the old data.

Craig Stark has an excellent article about setting GAIN and OFFSET - search for it on the internet.

I always advocate trying to run your imager at repeatable temperatures and to have taken Darks exposed to match your Lights and Flats. The other way to look at this issue is to decide on some common exposure times for your Lights and Flats and then you should always have Darks 'on file' to match these. For example I usually always expose either 300s or 600s subs, and these are always taken at a repeatable -15°C. So, I only need 300s and 600s Darks. For Flats, I get excellent results with 2s exposure times, so I keep a set 2s FlatDarks as well. I don't ever include Biases in my calibration data - I am using an imager that doesn't introduce bias noise that needs to be removed.

In your case, you have just complicated the whole process by using different Gains and Exposure times. Just remember to StarAlign before ImageIntegration (and to CFAdeBayer before StarAlign, if your imager needs this stage).

Hope this helps.

[niccoc1603]

Hi yes I know, that was done by mistake, so now I am wondering whether those two sets of data are unusable or I can still get something out of it

[John_Gill]

Hi,

This might not be ideal, but I do the following …   I process each stack of images to the point before doing a StarAlignment.  ie, Calibrated, CosmeticCorrection, Debayer (if required), SubframeSelector (I do not use Local Normalization).  Your image stacks are now "clean".

Now use StarAlignment to combine all clean image stacks.  Finish with ImageIntegration and DrizzleIntegration (optional).

space is not black
John

[niccoc1603]

Thanks John, that was my idea too

One thing, if I use SubframeSelector on each stack separately to weight the sub frames, am I correct in assuming than the weights will not be consistent across the stacks and so I can not use SSWEIGHT in ImageIntegration?
 

[John_Gill]

Hi,

The first SFS is used to find the best reference for each stack of images.  When combining stacks, I run another SFS on all images, in all stacks, to select the best image of everything.

John