Field curvature issues when merging data from 2 telescopes [Deep Sky] Processing techniques · Xanthus · ... · 5 · 405 · 5

Xanthus 0.00
...
· 
·  Share link
Hello, 

This is my first real post and well it's kind of a doozy.

I'm working on a project with a couple of other AP friends. We collectively gathered a bunch of data on M33.

I gathered Ha and Oiii data and my friends did a bunch of time using some OSC cameras with uv/ir filters.  We planned on using both sets for the RGB data. One telescope is a astrograph 6" newtonian, the other an askar FRA 500 refractor.

This is an alpha blend of both their data sets using Pixel math in pixinsight. They were star aligned prior.
ex1.jpg

Looks awesome right!?

Well.. 

In processing I'm seeing some major issues that I've spent a few weeks trying to combat.

The curvature issue seems to be most apparent in the corners. The stars are even separating at the very edges, like this top left corner.

I can take care of this by only using one person's star data however...

ex2.jpg

You can see this still affecting areas near the center of the galaxy, pulling them radially. Not as extreme but still detail loss compared to what I can get with just one of the 2 images. I cannot reliably remove these small star clusters of the galaxy even with starXtermintator so I'm at a loss on how to utilize both data sets without taking a detail hit.

ex3.jpg

This is the same area with just one of the telescopes. The stars don't pull to one side nearly as much. Both have stars like this so it's not guiding error on a singular scope either.

ex4.jpg

Anyone out there that has worked with multiple data sets have any suggestions on this?

Any help would be greatly appreciated!
Edited ...
Like
andreatax 9.89
...
· 
·  Share link
Well, I'd proceed as follows:

1. BXT on both sets
2. StarAlignment with 4th Order Surface Spline (or 5th or 6th if improvements can be found)
3. Merge back with Linear Fit applied beforehand, where applicable.
Like
GalacticRAVE 6.67
...
· 
·  Share link
Which pixinsight version are you using? In the newest one (1.8.9-2)  the star alignment routine + distortion correction has been overhauled substantially and provides much better/more stable results. If you use WBPP, beware that in the older PI versions, if  you get an error in case it cannot correct the image distortion, that error does not force WBPP to abort.

Matthias
Like
Xanthus 0.00
Topic starter
...
· 
·  Share link
I'm using the latest version on Pix.

I did try stacking both together with the raw data with wbpp and ended up with the same result. I'm not sure what distortion correction if any occurred though. I'll give that a check thanks.
Edited ...
Like
kaelig 1.81
...
· 
·  Share link
For myself, I use APP (Astro Pixel Processor). For me, it is the best for the Mosaic and the merge o different sources.
Like
Xanthus 0.00
Topic starter
...
· 
·  Share link
@GalacticRAVE@andrea tasselli

Thank you both so much. That really was what I was missing. I didn't really know about the different registration models and the surface spline options did the trick. Gonna experiment to see which looks the best but this resolves my major problem. 

For future onlookers in case this comes up. That is changed here in the star alignment tool.preview.jpg
Like
 
Register or login to create to post a reply.