StarNet processing issue + Focus and overexposed stars issue. - Forums

Topic starter

Aug 20, 2024

Hello everyone,

I'm new to the hobby of astrophotography. I've run into a problem and can't seem to find a solution.

I am using Siril and StarNet from inside Siril to separate stars. After the process is completed on the starless version, there are still stars. I've tried various stretching intensities, but the stars always remain.

This is an image before using StarNet:

This are the images of starless version and a starmask that are generated:

Related to the photography process itself, exposures of 180 seconds and 100 frames were used. I think that the stars in the photo are overexposed and that the focus is not good, but I don't have enough experience to be sure since this is my first photo.

With the data i managed to get this final result:

Thank you for all the advice in advance!

Kind Regards!

Respectful Engaging

Tony Gondola

Frequent contributor

Aug 20, 2024

Make sure that you are running starnet before you do any stretching and make sure you tell starnet to pre-stretch the image during extraction. There is also a command in the stars menu that will de-saturate your stars for you. I would run that first.

Helpful Concise

NeilM

Aug 22, 2024

I agree 100% with Gondola. Focus and star 'bloat' (saturation) are definitely issues with your image, but Starnet is pretty forgiving with those. Focus issues can cause haloes with Starnet, and that doesn't appear to be a problem in your starless image. The good news is that for your 180 second exposures your starts are perfectly round which means that you nailed the polar alignment and guiding - well done. Focus should be a relatively easy thing to fix in the future.

I took your low quality 400k jpeg file and did what Gondola suggested and desaturated the stars before running Starnet. The result is below. As you can see essentially all stars have been eliminated.

There's some nice detail there.. with this data I think you should be able to create a great image.

Good luck

Neil

Helpful Respectful Engaging Supportive

JH.Astrophotography

Topic starter

Aug 22, 2024

Make sure that you are running starnet before you do any stretching and make sure you tell starnet to pre-stretch the image during extraction. There is also a command in the stars menu that will de-saturate your stars for you. I would run that first.

Thank you very much for the advice. I tried what you suggested and it worked much better.

I agree 100% with Gondola. Focus and star 'bloat' (saturation) are definitely issues with your image, but Starnet is pretty forgiving with those. Focus issues can cause haloes with Starnet, and that doesn't appear to be a problem in your starless image. The good news is that for your 180 second exposures your starts are perfectly round which means that you nailed the polar alignment and guiding - well done. Focus should be a relatively easy thing to fix in the future.

I took your low quality 400k jpeg file and did what Gondola suggested and desaturated the stars before running Starnet. The result is below. As you can see essentially all stars have been eliminated.

There's some nice detail there.. with this data I think you should be able to create a great image.

Good luck

Neil

Thank you very much for the input. I was looking photos of this object that other people did and it seems so much sharper than I could accomplish here. I guess that is a focus issue as you said (and I assume additionally better processing techniques).

Respectful Supportive

NeilM

Aug 22, 2024

And then just for fun I spent a few minutes adjusting the curves of the starless image, removing noise and then recombining it with the slightly less saturated star mask. This is all very poor low quality data from your original jpeg that you posted, so you should be able to achieve fantastic results with your raw data.

JH.Astrophotography

Topic starter

Aug 22, 2024

And then just for fun I spent a few minutes adjusting the curves of the starless image, removing noise and then recombining it with the slightly less saturated star mask. This is all very poor low quality data from your original jpeg that you posted, so you should be able to achieve fantastic results with your raw data.

Nice result with that jpeg.

I guess learning curve for great processing is pretty steep. Whatever I tried I couldn't really get the result I am satisfied with and I expected more details with that much integration (telescope is 80 mm refractor at f6.4).

NeilM

Aug 22, 2024

I'll add that I think you do have a lot of detail in your data! As you say, it will take a little experimentation to draw it all out.

Well Written Respectful

mlewis

Aug 28, 2024

I think your image is a great first effort - not that bad at all.

As far as overexposed stars go, I like to take separate data for the stars with shorter exposure times (and sometimes maybe lower gain too). This is easy and quick to do. I usually take between 10 and 30 minutes of data, usually 30 second subs or even shorter depending on how fast your setup is. I then stack and stretch, using a very moderate stretch to keep from overexposing most of the stars, and then add some saturation. This then allows you to use that star data when you add your stars back in, giving you tighter stars, with less bloat and more color. It only adds a few minutes to both the collection and post processing, and can really improve the look of the final result IMHO.

ML

Helpful Respectful Engaging Supportive

JH.Astrophotography

Topic starter

Dec 30, 2024

Just a short update on this topic for all those who may encounter the same problem in the future.

The main problem that caused overexposed and blown out stars was the use of an astro camera without a UV/IR cut filter.

The pictures also show a pink hue around the stars, which is caused by light not in the visible spectrum, this hue is difficult to remove and due to the different focus point of the invisible spectrum of light compared to the visible spectrum, problems with focus occur.

Using the UV/IR cut filter solved all my problems listed above in this thread.

Helpful Concise