Combining Datasets from different chips in PixInsight (WBPP)

[Derek Warltier]

I just watched Nik Szymanek's tutorial on combining datasets from different telescopes in APP, but I am looking for a similar path using PixInsight.  I am trying to process NGC 1291 using the CHI-1-CMOS dataset, but there is very little colour in the image (apart from the stars).  I decided to combine the dataset from CHI-1-CCD  and also some other data I had from Chilescope.  WBPP could not process the CCD data (Bin 1) with the CMOS data (Bin 2) and I ended up with multiple masters. I adjusted the exposure tolerance to combine the different exposure lengths, but could not get WBPP to combine the data for different binning.  Any suggestions?

[Astro Newbie]

20 hours ago, Derek Warltier said:

I just watched Nik Szymanek's tutorial on combining datasets from different telescopes in APP, but I am looking for a similar path using PixInsight.  I am trying to process NGC 1291 using the CHI-1-CMOS dataset, but there is very little colour in the image (apart from the stars).  I decided to combine the dataset from CHI-1-CCD  and also some other data I had from Chilescope.  WBPP could not process the CCD data (Bin 1) with the CMOS data (Bin 2) and I ended up with multiple masters. I adjusted the exposure tolerance to combine the different exposure lengths, but could not get WBPP to combine the data for different binning.  Any suggestions?

Hi Derek,

It doesn't look like there is any obvious way within PixInsight other than trying to align the masters and combine them manually. 

Sometimes Star Alignment fails, for any number of reasons, but you could try aligning all the master stacks of one set to one of the master stacks of the other camera. The CMOS camera has a higher resolution so you may probably want to use its Luminance stack to align the CCD stacks. As a result of the upscaling, the aligned CCD stacks may well be noisier. To reduce problems from gradient differences between the stacks maybe first correct each stack on both camera sets prior to aligning.

Hopefully the newly aligned stacks will be compatible but the next thing is how best to combine them. Using Pixel Math to add each pair together you would need to find a way to weight them, not something I know how to. Using Image Integration manually you can add 2 sets of each master stack, since it will not allow less than 3 sets. Regarding the criteria to use to weight them I'm not sure, maybe SNR. 

This ought to be a quick but rather untechnical way to do it but hopefully won't take up too much of your time. It's even possible that the resulting masters might be worse than the individual ones, so be prepared! Many apologies if you've tried this and it failed, this is just a theoretical solution.

Hoping that soon someone else will 'chip in'.

CS Ray

[Reggie Jones]

Hi Derek.  Not sure there is a clean way to do this in PixInsight using WBPP.  You could try downscaling the CCD FITS data to 2x2 binning before using WBPP to process the data to see if that would work.  I'll see if Adam Block has anything on doing something like this and if so, share it here.

Reggie

[Kurt Lanes]

Jarmo Ruuth has a PI script that I use a lot. You can combine frames either masters of subframes and you can have only 2 frames if you like. The script is called AutoIntegrate and is available as a repository here https://ruuth.xyz/autointegrate/.

 

Kurt

[Alexander Curry]

We also have an excellent tutorial by Jarmo on using AutoIntegrate:

https://telescope.live/tutorials/how-easily-automate-your-pixinsight-integration-workflow-autointegrate

[Kurt Lanes]

Thanks Alexander I will check that out. I use it all the time.

As I understand it, you need to use align by coordinate. I am running a batch thru now using SPA-1 and SPA-2 data. See this post. I will try an post alink here to the PI froum if it will let me.

 

 

Kurt

[Kurt Lanes]

On 3/22/2024 at 3:32 PM, Derek Warltier said:

I just watched Nik Szymanek's tutorial on combining datasets from different telescopes in APP, but I am looking for a similar path using PixInsight.  I am trying to process NGC 1291 using the CHI-1-CMOS dataset, but there is very little colour in the image (apart from the stars).  I decided to combine the dataset from CHI-1-CCD  and also some other data I had from Chilescope.  WBPP could not process the CCD data (Bin 1) with the CMOS data (Bin 2) and I ended up with multiple masters. I adjusted the exposure tolerance to combine the different exposure lengths, but could not get WBPP to combine the data for different binning.  Any suggestions?

Derek

I just finished combining SPA-1-CCD, and SPA-1-CMOS with my TEC140FL data (0.9" IFOV) using the Image Analysis -> ImageSolver script and utilities->AlignByCoordinates script. Load all the subframes into the ImageSolver and create WCS files then choose a reference file and run AlignByCoordinates. In my case I choose the TEC140 with 0.9" pixels and registered this to the larger 2.02" and 4.75" files. I am now tried the utilities->FFTRegistration script and aligned them and integrated w/o generating th ealigend files.

 

Kurt