Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

NGC7822 - original PixelMath palette.

Adreneline

By Adreneline
in Imaging - Deep Sky

 Share

Recommended Posts

Adreneline

Adreneline

Posted
Posted (edited)

Decided to experiment with PixelMath with Ha, OIII and SII - nothing too avant garde.

7822_clone-DeNoiseAI-denoise-DeNoiseAI-denoise.thumb.png.beb5556268a3bdfaf9088c3bbd24834b.png

24 x 180s of Ha, OIII and SII using a Samyang 135mm + ASI1600, unguided on iOptron CME25-EC.

Processed in PI - first use of a trial of Topaz DeNoise AI.

Thanks for looking.

Adrian

Edited by Adreneline
Typo
  • Like 11
Link to comment
Share on other sites
Adreneline

Adreneline

Posted
  • Author
Posted
3 minutes ago, Kinch said:

The DeNoise certainly works.......perhaps a bit too well for Astro.

Thank you for the observation Brendan. I've not decided yet whether or not to buy in 28 days time. The software carries out a combined denoise and sharpen each set independently with a slider control. I need to experiment a lot more and really appreciate your feedback. I had DeNoise and Sharpen both set at 15 (out of 100) so a fairly minimal amount and I was quite pleased with the result but maybe I'll pull back the DeNoise a little and keep the Sharpen at the same value.

Over the three nights of data collection conditions were not ideal. I notice that whenever I decide to collect OIII the conditions go down the pan!

Hoping to move house soon to a Bortle 3/4 region from the current Bortle 5/6 and I'm thinking I might scrap all my data and start again. A new beginning!

Thanks again.

Adrian

Link to comment
Share on other sites
Adreneline

Adreneline

Posted
  • Author
Posted
4 minutes ago, MarkAR said:

I think with some processes it's too easy to go over the top,

I am concerned I have gone "over the top" with this image - I tend to prefer a much more pastel palette but I thought I might push this one to see how the noise reduction coped.

Glad you like the colours.

Adrian

Link to comment
Share on other sites
Kinch

Kinch

Posted
Posted
22 minutes ago, Adreneline said:

Hoping to move house soon to a Bortle 3/4 region from the current Bortle 5/6 and I'm thinking I might scrap all my data and start again.

Good luck with the move - I would hate the thoughts of doing that again....I have lived in many houses over the last 40 years. No need to scrap all your data though.....hard drive spave is relatively cheap 🤓.

Re Marks comment....."sometimes repeating an operation at much reduced settings can work a lot better".....I have actually got into a habbit now of going 'over the top' on some processes but then mixing the original with the 2nd rendition.....sometimes I find a nice balance that way (though sometimes not 😟)

  • Like 1
Link to comment
Share on other sites
MarkAR

MarkAR

Posted
Posted
1 minute ago, Kinch said:

.I have actually got into a habbit now of going 'over the top' on some processes but then mixing the original with the 2nd rendition.....sometimes I find a nice balance that way (though sometimes not 😟)

That can actually work very well, not all the times though as you've found out. It's all trial and error to see what works and it can be very different for another similar target.

  • Like 2
Link to comment
Share on other sites
MarkAR

MarkAR

Posted
Posted
11 minutes ago, Adreneline said:

I am concerned I have gone "over the top" with this image - I tend to prefer a much more pastel palette but I thought I might push this one to see how the noise reduction coped.

Glad you like the colours.

Adrian

I don't think it's OTT at all colour wise, pixel peeping does show blotchy artefacts so maybe just dial down the Topaz Ai a bit.

  • Like 1
Link to comment
Share on other sites
AbsolutelyN

AbsolutelyN

Posted
Posted

Great image, personally like the second version best. Much brighter nebulosity works much better for me. I've also recently taken out Topaz trial. I've mostly used Imagenomic Noiseware on astro but Topaz gives far smoother results. Used carefully it's certainly a very useful tool. Pretty certain I'll buy it when I can afford but might wait for a discount code.  

Link to comment
Share on other sites
Adreneline

Adreneline

Posted
  • Author
Posted
2 minutes ago, AbsolutelyN said:

Great image, personally like the second version best. Much brighter nebulosity works much better for me. I've also recently taken out Topaz trial. I've mostly used Imagenomic Noiseware on astro but Topaz gives far smoother results. Used carefully it's certainly a very useful tool. Pretty certain I'll buy it when I can afford but might wait for a discount code.

Thank you Tristan.

I used Dfine within DxO Photolab to reduce just the noise on the second image which I've used for ages as a Photoshop plugin but therein lies another story!

I am looking for an alternative to PS. I bought PS Extended back in about 2012 but it is no longer supported by Adobe (no RAW support for my new M6 MkII) and worse than that it won't run or even install on MacOS Catalina and won't install properly on Mojava. As the latest PixInsight requires an OS above High Sierra I'm stuck and I refuse to pay the Adobe subscription when I purchased a full version.

So I've been experimenting with Luminar 2018 and with Topaz DeNoise AI. I also downloaded DxO (which I'm still trying to get my head around); the advantage being I can use my DxO plugins. I could also use the DxO plugins if I upgrade my old Luminar 2018 to the new Luminar 4 (rumour has it Luminar AI is on the horizon). Too much choice! I do like Dfine so I am very tempted to stick with it but thought I'd give Topaz as well. I've got some low-light terestial photos and Topaz has worked wonders on those with the combined DeNoise and Sharpen.

Too much choice!

Adrian

 

Link to comment
Share on other sites
Adreneline

Adreneline

Posted
  • Author
Posted
3 minutes ago, wimvb said:

Doesn’t mmt (multiscale median transform) in pixinsight do the same?

Hi Wim. To be honest I've only used MLT and I've had mixed results on LRGB/NB channel images; it has worked better for me on osc/dslr data.

Perhaps I should investigate MMT as well.

Adrian

  • Like 1
Link to comment
Share on other sites
Kinch

Kinch

Posted
Posted
1 hour ago, Adreneline said:

Just wondering @Kinch if this version looks any better to your eyes?

For me the actual background is better in that it is not looking 'wiped clean' like the original. But then some other things are happening and although the nebulosity is brighter (better) is is somehow less sharp and that part has been smoothed out now. 

From above posts I note you have PixInsight. I use TGV denoise in that programme with good success. The trick I find is to use it more than once in the workflow (say on each master....then again on the final) but very use a lower strength with more iterations. Do it on apreviews for several parts of the image - to check how things will turn out overall.....because when doing it 'for real' with many iterations, it will take a long time on the actual full image.

  • Thanks 1
Link to comment
Share on other sites
TerryMcK

TerryMcK

Posted
Posted (edited)

I have Topaz Denoise and personally think it is very good. However it is possible to go over the top with the sharpening function especially in auto mode. So best to tackle each image subjectively rather than use the same settings everytime

Edited by TerryMcK
  • Thanks 1
Link to comment
Share on other sites
wimvb

wimvb

Posted
Posted
On 11/09/2020 at 14:53, Adreneline said:

This is the image with MMT applied. Not sure I've used the optinum settings so any advice would be appreciated.

Using a process to reduce noise and sharpen details at the same time, seems counterintuitive. Noise reduction is done on the darker, low snr areas, while sharpening is done on the medium and bright areas. But it works.

In your image, the smaller stars are blocky (undersampling? or a starmask that masked too little?) This can also be the result of deconvolution. For deconvolution to work, the image needs to be oversampled, rather than undersampled.

To reduce noise further, try mmt on chrominance only, but quite aggressively on up to 8 layers. With a mask that protects the stars and nebula, this neutralises the background colour noise, but preserves all the fine detail.

https://jonrista.com/the-astrophotographers-guide/pixinsights/effective-noise-reduction-part-2/

  • Thanks 2
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.