Donuts like stars

[whiteLion]

Hi guys!
Well, I got this problem, and I saw it after preprocessing. The donut like stars are only on Red channel.
It is from poor focus or something else?
I split channels, extract L and apply STF to the same preview.

[whiteLion]

here are and the other channels plus the RGB file.
All are the same preview, the same area, same stars.

[Dimitris Platis]

Yes...its probably unfocused..

[whiteLion]

And is there an solution to fix this?

[Dimitris Platis]

U can make a star mask and apply it to RGB and use convolution to blur the center.....or just do that in the R channel

[Altayer]

make sure the the filter is tight on the filter disk.

there is a rotate-to lock mechanism on round filters. if it is loos, it will change position up or down between the clearance of the filter. and

results in change of focus.

you can notice this if the focus changes significantly between the the first and the the last frame of the red exposures.


Regards

[MikeOates]

All the images you show above are unfortunately out of focus, the red channel just being more out than the others. As Dimitris says you can improve the donut shape of the red channel by using convolution through a star mask. You can also try using deconvolution and Morphological Transformation to reduce the star sizes, (see links at the end).

But to be honest the only real solution to fix this is to re-shoot the sub frames again but taking great care to get the focus right. Sorry if that sounds blunt, I don’t mean to sound harsh, but focusing is very important and can be a difficult, sometimes very time consuming process, but extra time taken will reap rewards in your final image. Not only will it look better, but you will be able to resolve fainter stars and get more detail in nebulae, galaxies etc. Remember to keep checking and adjusting the focus throughout the imaging session as the focus will change depending on temperature changes and how your scope is constructed.

I am assuming the data was captured with a colour camera? If that’s the case, what your seeing is the red part of the spectrum focusing slightly different to the other channels. The closer to the focus you get, the less of a difference will be seen.

I hope that helps.

Mike

Some links:

Deconvolution example:
http://www.pixinsight.com/examples/M81M82/

Harry's video (Star mask1) shows you how to make masks and the use of Morphological Transformation:
http://harrysastroshed.com/pixinsight/pixinsight%20video%20html/pixinsighthomeinter.html

[oldwexi]

And is there an solution to fix this?

Yes, reshoot the Images now properly focused, and dump the old ones
in the "Focus troubles experienced" Folder.

Gerald

[MortenBalling]

I'm not sure that you need to throw away your work. Donut like stars like these can i some cases be restored by using MorphologicalTransform carefully. The trick is not to overdo it, and to use a good starmask.

Generally the subjective visual sharpness of an astrophoto can be divided into the stars, and all the rest. A large portion of the percepted sharpness lies in the stars. Nebulae etc. are much less critical when you look at the final image. Furthermore, you can try to deconvolve the nebulae and galaxies using a sampled PSF, or even one that you make yourself by experimenting with the dials.

Nothing beats focusing the telescope, but many of the tools available in PixInsight are designed to fix a "not so perfect" image.

Best

Morten