Problem with Star Aligbnment in PixInsight - Forums

Topic starter

Mar 17, 2023 · Edited: Mar 17, 2023

I have 54 four- minute subs of M81 that I am trying to align, taken at a focal length of 1422mm.

Many of the subs are failing to register. Which parameters in StarAlignment are most likely to help me align the maximum number of subs?
And should I try using a difference reference sub?

Here are the error messages the process console is reporting:

"Unable to find an initial set of putative star pair matches" and it also says:

"0 star pair matches found - need at least six matched stars"

Thanks so much,
Jerry

Well Written Engaging

andrea tasselli

Frequent contributor

Mar 17, 2023

If they are failing to register and there are more than 6 stars in each of them and the frame shifts are minor across the board then the issue might be whether they are in focus or not. If you click on the documentation link there is a good explanation of the various parameters.

Markus R.

Mar 17, 2023 · Edited: Mar 17, 2023

Did you debayer the Subs before trying to align them? If not, that could be the problem.

Well Written

Jerry Gerber

Topic starter

Mar 17, 2023

I got help as to how to set the StarAlignment settings to handle elongated stars. I've got all the subs registered now but I can't say I am happy with the quality of the photo…

Jonathan W MacCollum

Mar 18, 2023

There are many possible reasons for this, but one possible case where this can happen is if you have "Restrict To Previews" checked under the "Star Matching" section, and you are trying to align the targets to a view that has preview windows on areas with little-to-no stars in it (for example if you defined preview windows on a background reference to little to no stars.)

If this happen to be the case for you, uncheck Restrict to Previews to allow alignment to use the entire image.

Cheers

Helpful

Steve Solon

Mar 18, 2023

Hi Jerry,
I assume all images are properly calibrated. If you're using a CMOS camera, the problem could be with hot pixels, even if you've used CosmeticCorrection. I had the same issue with images taken at 2800mm. There were very few stars in the field I was shooting, but I thought there were still enough to align with. A video by Adam Block indicated that leftover hot pixels would be causing the alignment algorithm to fail.

If the above is the case, try Debayering the subs again, only this time, try using SuperPixel as the Demosaicing method instead of VNG. This helps to 'flatten' out the signal of hot pixels so they're not as noticeable. If not, then, as mentioned above, try varying the sensitivity parameters, etc. until StarAlignment recognizes the stars it wants to align.

Hope this helps a bit.
- - Steve

Well Written Helpful

Jerry Gerber

Topic starter

Mar 18, 2023

Hi Jerry,
I assume all images are properly calibrated. If you're using a CMOS camera, the problem could be with hot pixels, even if you've used CosmeticCorrection. I had the same issue with images taken at 2800mm. There were very few stars in the field I was shooting, but I thought there were still enough to align with. A video by Adam Block indicated that leftover hot pixels would be causing the alignment algorithm to fail.

If the above is the case, try Debayering the subs again, only this time, try using SuperPixel as the Demosaicing method instead of VNG. This helps to 'flatten' out the signal of hot pixels so they're not as noticeable. If not, then, as mentioned above, try varying the sensitivity parameters, etc. until StarAlignment recognizes the stars it wants to align.

Hope this helps a bit.
- - Steve

Hi Steve,

I think the problem is rooted in collimation. I was using my Edge 8" and the collimation was so-so. I noticed many of the stars were elongated, and very large as well. I read that PI's StarAlign routine requires some settings to be changed to deal with elongation. I changed those and all 54 subs succeeded in registering.

But the image is really mediocre, not because of PI but because I've to to get outside and collimate soon. I just bought a Geoptik artificial star with a small, 50 micron size star which should make collimation easier. Once I get the stars looking better I think PI will have no problem registering them. I've never had an issue with my refractor, which has pinpoint, really sharp stars...

Jerry

Helpful Insightful Respectful Engaging Supportive