Last nights efforts

[Gavin1234]

Finally got some clear skies, sorted out my new alignment issues and got a couple of images last night.

These have just had a quick stretch so far because I had to leave for Melbourne this morning. A lot more noise than I would have thought and I need to remove the rings from the stars in the helix picture amongst plenty of other things.

I’ve settled on gain of 160 and 5 min subs for both of these targets as a standard approach. I feel like only now after months of learning that I’m consistent enough to really start building up data to eventually end up with a good image.

The horsehead is about 15 x 5 min subs and the helix is about 40 x 5 min subs. I’ve used 250 bias, 50 darks and 50 flats. Calibrated and stacked to masters in PI. Any thoughts on why I would have so much noise in the picture at this stage? I would have thought that with 40 x 5 min subs it should be much cleaner at this point. Possibly a calibration, integration issue? I’m just learning that process with PixInsight at the moment.

[cometcatcher]

It's possible to over stretch any image, no matter the amount of data. The Helix has some processing artifacts. There are rings around the stars.

That's a really nice horsehead region by the way.

[ChrisV]

+1 Really nice images.
Do you know where the rings appeared when processing in PI? Might have been when doing noise reduction?

[Gavin1234]

Yeah I think it was tvg noise reduction or something similar. When I get home I’ll re run it with the “de-ring” option selected.

Just curious as to why there is so much noise to begin with.

[Atmos]

Your horsehead is particularly good, I’d be happy with that :)

Noise is a difficult one as it does depend on what you are photographing. 40x5 is 3.3 hours so in many ways it isn’t a lot of data and when you are photographing faint areas (fainter nebulosity around the horsehead), it will require a lot more data.

Other aspects are cooling, good darks and flats and probably more importantly, dithering. Dithering makes a big difference as it allows you to remove FPN which is a big contributer to noise.

[ChrisV]

If its the TGV denoise , from the defaults try altering the edge protection. I find I have to change it ask the time. And have a look at Harry's Astroshed video etc. Trouble is everywhere you look does is differently.

I've also been doing the Helix. It's not very bright so need a lot of data to get SNR to an 'acceptable' level. I've found it difficult.

[ChrisV]

Colin, I've tried dithering and found it good - although the backlash on my Dec axis doesn't think so. How much should you dither by ? I've done 3 pixels ?? Does it depend on how many subs you go for ?

[Atmos]

I find myself always having to change TGVDenoise as the standard -3 is far too high. I usually start by dropping it to -7 and then slowly raise it but never go passed -6.

I usually use 3 pixel dither as it does do fraction of pixel movements.

[xelasnave]

Great stuff Gavin.
I will bet you can get that noise out...
I did an old Horse Head the other night which was so bad I never bothered up till now to process it and it did not look much until the final noise clean up in startools and you can bet Pixinsite would do even better.
As to dithering ...the way my polar was off and short subs that worked as the object would move maybe 10% of the frame in an hour session☺ auto dither.
I cant wait to try it this year and Orion as well...

Alex

[Gavin1234]

Thanks guys. I’m looking forward to getting back to Sydney tonight so I can keep working on these. I’m going to set aside a lot of time to look at and experiment with different methods for noise reduction.

Dithering is also something I haven’t tried yet but it was next on my list. Next time I get clear skies on a weekend I will try it at -7 I think. I’ll need a weekend so I have time to setup a really good polar alignment.

It’s all good practice while I’m building up data on these subjects.