Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

The Smokin' Hot Galaxy - M82 returns - this time properly smokin'


powerlord

Recommended Posts

Apologies - this is cross posted to the galaxy comp thread, but comments not allowed there and I wanted some feedback, to cross posting here.

I've been doing this just over 2 years now, and tried imaging Bode's Galaxies a few times. However I've never really been happy with the results. I think most of it comes down to impatience. If after a night of imaging I don't see something good that either looks complete or at least good enough to continue I tend to give up.

And Bode's Galaxies are weird like that - they come out very bright - visible in even a single sub. But lacking in detail, very red/white and boring. I've never faced up the challenge of extracting detail from that bright core.

But, it's galaxy season, and with my 'Iraqi Supergun' SW 300PDS I was looking for targets near Polaris so that movement would be minimised to help this beast stay guiding at a decent rate on my EQ6-R, and without bumping into either the sides of the new observatory. So  started having another go at M82 - with my asi2600 and Ha/Oiii via the L-ultimate. In fact I posted my first (red/yellow) attempt on imaging/deep sky last week. Olly pointed out that it really shouldn't be that colour... and yeh that got me thinking I need a new approach here!

Anyway, with terrible weather, it's taken me quite a while just to gather this much data so I think for now, this is it.

In the end, mostly 3 minute subs for the RGB (totalling 7 hours), 10 minute subs for the Ha/Oiii (totalling 4 hours), and if I'm honest probably 6+ hours of doing, redoing, doing and redoing the image processing before I was half way happy with it!

After the feedback from Olly, I went back to the  beginning with my RGB data - and this time starting off by letting Siril photometrically colour correct it to get me in the right ballpark.

And instead of getting to the end and then adding the Ha, after a lot of experimentation - I ended up using  pixelmath in Siril to blend it in from the start, along with my Oiii data from the L-ultimate: I created R1 by creating a 0.4*R+0.6*Ha, then R2 as 0.4*R1+0.6*Ha. I then used R2 as red. For the green I just used green, and for the blue I used the stacked blue/green from my L-ultimate.
 
The result immediately started to look good, and it was then into Affinity Photo for editing, also making use of Russell Croman's plugins (starXterminator and noiseXterminator).

Like every astrophoto ever, it could do with more integration time, but I'm now unlikely to get it this season - so here it is.


stu

M82-final.thumb.jpg.2cec2ca438822dddc096a0e5d7c0c14f.jpg

And one with more NoiseX and a bit less red flare... better ? worse ?

m82finalv2.thumb.jpg.5ccd73375554ac69d77021702d34f8ae.jpg

 

Edited by powerlord
  • Like 35
Link to comment
Share on other sites

No ai tools used at all vlaiv. Other than noisex, which doesn't make things up so not sure what your referring to ? Unless it's the extended red, which I'm having trouble removing - tried in the noisex version.

Also, I compared very carefully with hubble, and as I say, other than my lack of managing to clean down the edges of the ha its identical ?

  • Like 2
Link to comment
Share on other sites

hubblevme.gif.044a846d319e5db6a433e98c88eacd0c.gif

That's vs Hubble. And as I say, the only thing I'm struggling with it getting rid of the red background extending from/around the Ha - but I didn't want to loose the fine bits I've got - though I did try in the noiseX second version, trying to mask and remove the red background noise a bit.

Edited by powerlord
  • Like 6
Link to comment
Share on other sites

Overall I think it's very good. The only bit that jars is the brown region which surrounds the Ha jets. I think that is spurious but, for the rest, you have good agreement with Hubble. The main outer glow may also be a bit blue but, hey, the image has something to say.

I have to say that all these complicated ways of adding Ha leave me a bit mystified because, in Photoshop, it is so easy to control where the Ha is and is not in play. Call that the last gasp of a dinosaur.

:grin:lly

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Yeh that's the bit  I'm referring to. I'll try again.

It's same in affinity I'm sure, I just need to walk back some layers. My problem is I often edit remotely , and never learn that fine stuff like this never views well over vnc.

Thanks.

Stu

 

Link to comment
Share on other sites

26 minutes ago, powerlord said:

No ai tools used at all vlaiv. Other than noisex, which doesn't make things up so not sure what your referring to ?

image.png.d0c6c91ad32310fc95c6950f98fe3bf6.png

I added red emphasis - that is screen shot from NoiseX.

It is neural network that is trained to "guess" what the actual image is given the noisy version (it sort of makes up the features).

Here are a few examples I'm referring to:

image.png.43b317b0f0977209a0f119a0bb704867.png

Top is Hubble reference, and bottom is your image. Two stars in top reference image have no Ha signal around them. In bottom - AI tool mistook noise around stars for Ha signal and "connected" it with other "Ha signal" that is in the area - all of which are noise level and not true Ha signal if we look at above reference image.

Then there is this "web" of features - that does not look like that in reference image:

image.png.96e054aba35aeb22cbee9dd2d36ab1f1.png

There is no central ring like structure with 5 "spikes" coming out of it - nothing like that in reference image.

image.png.f7b72b8820e4265b5234a381c7d2fb40.png

then subtle variations in what appear to be bright and dark regions in Ha

 

Link to comment
Share on other sites

🙄

the ORIGINAL stretched stacked image.  Oh look - a 'web' of features. Sorry my skills are not quite up to Hubble's.

Screenshot2023-04-25at18_44_08.thumb.png.35ac319f669062e44a1c778d217870f4.png

Oh,, and the 'Ha signal around them' - that's because I stacked the Ha as I said I did - and my guiding last night was terrible - it didn't affect the dim stuff as it got back on target but it did mean my Ha stars are rubbish. Hence some bright ones have a sort of Ha halo (like the one on the right above) - no AI bogie men for you to worry your head about.

It's almost like, I dunno, I'm imaging it through 100 miles of turbulent atmosphere, and subject to limited seeing and general guiding issues and all the rest of the regular stuff we all are . jeez give me a break.

Edited by powerlord
  • Like 4
Link to comment
Share on other sites

30 minutes ago, powerlord said:

jeez give me a break.

I just wanted to point out that if you image at such pixel scale - features that are below resolving capability of a system on a given night will be made up to look like something that is not there if you use AI tools (and sometimes even if you are not below resolving capability of the system).

If you look at the size of that "ring" structure - it is about the same size as a pair of stars in the left part of the image that are not resolved properly.

Once image is shown at resolution that it supported by the data - very little difference between it and reference image:

image.png.1e88b127d2e89278781637dc74b5e4a4.png

 

Link to comment
Share on other sites

2 minutes ago, cuivenion said:

Couldn’t normal sharpening tools cause these artifacts as well though?

There are two things that can happen when using "normal" sharpening tool that distort the image.

1. Increase in noise levels

Sharpening boosts high frequency components of the image, but noise is spread over all frequency components so part of the noise that is in high frequency range will be boosted as well (we can't effectively separate the two - otherwise denoising would be trivial affair)

2. Inaccurate blur kernel assumption

In order to effectively sharpen the image - blur that was applied to the image must be guessed accurately - otherwise additional distortion can happen.

For most astronomical images Gaussian type blur is very good guess and not much issues arise if one uses that as blur kernel. FWHM of that blur kernel is also imprinted in the image - it is star FWHM.

In that sense - with good sharpening tool - there is little that can go wrong except for "excessive use of the tool" and the fact that it creates additional noise.

Link to comment
Share on other sites

2 minutes ago, cuivenion said:

Are you saying don’t use AI sharpening if the image is over sampled?

I would personally avoid using AI tools completely, but that is just my view on things.

Link to comment
Share on other sites

4 hours ago, powerlord said:

Apologies - this is cross posted to the galaxy comp thread, but comments not allowed there and I wanted some feedback, to cross posting here.

I've been doing this just over 2 years now, and tried imaging Bode's Galaxies a few times. However I've never really been happy with the results. I think most of it comes down to impatience. If after a night of imaging I don't see something good that either looks complete or at least good enough to continue I tend to give up.

And Bode's Galaxies are weird like that - they come out very bright - visible in even a single sub. But lacking in detail, very red/white and boring. I've never faced up the challenge of extracting detail from that bright core.

But, it's galaxy season, and with my 'Iraqi Supergun' SW 300PDS I was looking for targets near Polaris so that movement would be minimised to help this beast stay guiding at a decent rate on my EQ6-R, and without bumping into either the sides of the new observatory. So  started having another go at M82 - with my asi2600 and Ha/Oiii via the L-ultimate. In fact I posted my first (red/yellow) attempt on imaging/deep sky last week. Olly pointed out that it really shouldn't be that colour... and yeh that got me thinking I need a new approach here!

Anyway, with terrible weather, it's taken me quite a while just to gather this much data so I think for now, this is it.

In the end, mostly 3 minute subs for the RGB (totalling 7 hours), 10 minute subs for the Ha/Oiii (totalling 4 hours), and if I'm honest probably 6+ hours of doing, redoing, doing and redoing the image processing before I was half way happy with it!

After the feedback from Olly, I went back to the  beginning with my RGB data - and this time starting off by letting Siril photometrically colour correct it to get me in the right ballpark.

And instead of getting to the end and then adding the Ha, after a lot of experimentation - I ended up using  pixelmath in Siril to blend it in from the start, along with my Oiii data from the L-ultimate: I created R1 by creating a 0.4*R+0.6*Ha, then R2 as 0.4*R1+0.6*Ha. I then used R2 as red. For the green I just used green, and for the blue I used the stacked blue/green from my L-ultimate.
 
The result immediately started to look good, and it was then into Affinity Photo for editing, also making use of Russell Croman's plugins (starXterminator and noiseXterminator).

Like every astrophoto ever, it could do with more integration time, but I'm now unlikely to get it this season - so here it is.


stu

M82-final.thumb.jpg.2cec2ca438822dddc096a0e5d7c0c14f.jpg

And one with more NoiseX and a bit less red flare... better ? worse ?

m82finalv2.thumb.jpg.5ccd73375554ac69d77021702d34f8ae.jpg

 

Excellent image! On my Galaxy Tab, v 1 looks "punchier". So, please don't enter it in the competition section. :D

Btw, R1 = 0.4×R + 0.6×Ha and R2 = 0.4×R1 + 0.6×Ha. Doesn't that just mean that R2 = 0.16×R + 0.84×Ha, so 84% Ha and 16% red? In other words, will you get the same result with just one pixelmath expression?

  • Haha 1
Link to comment
Share on other sites

I counsel peace!

For me, the X suite AI tools are excellent but not perfect.  Some vigilance is called for to check that they don't make spurious additions or deletions. That means checking them carefully against the unmodified image at the same stretch. I now do this. Rejecting them altogether would be like throwing the baby out with the bathwater, for me.

When I add Ha to red in Photoshop (and since Ha is red, that's where I put it) I do so by making it a layer on top of red and choosing blend mode lighten. This means the Ha is only applied where it is brighter than the red.  However, at this stage I still have fine control of the mathematics involved in the blend. I can bring in the black point of the Ha and I can shape the curve in Curves. Simply by shaping the curve I'm creating new formulae adapted to the image in question. The mathematics of the formulae are invisible to me but the consequences of the formula are visible in real time in front of my eyes. Why would I do it any other way?  In Powerlord's image I would be able to look at those magnificent Ha jets (of which I'm very jealous! 😄) and shape the curve so that only the thin jet filaments made it above the red floor.

The Spanish Inquisition will accuse me of heresy or 'painting,' to which I can only say that PI users will do exactly what I do in the end and that is choose the final result based on subjective preference.

Cheating? Where are the rules?  We make our own. My rule is to try to make beautiful and informative pictures.

Olly

Edited by ollypenrice
  • Like 5
Link to comment
Share on other sites

15 minutes ago, ollypenrice said:

I counsel peace!

For me, the X suite AI tools are excellent but not perfect.  Some vigilance is called for to check that they don't make spurious additions or deletions. That means checking them carefully against the unmodified image at the same stretch. I now do this. Rejecting them altogether would be like throwing the baby out with the bathwater, for me.

When I add Ha to red in Photoshop (and since Ha is red, that's where I put it) I do so by making it a layer on top of red and choosing blend mode lighten. This means the Ha is only applied where it is brighter than the red.  However, at this stage I still have fine control of the mathematics involved in the blend. I can bring in the black point of the Ha and I can shape the curve in Curves. Simply by shaping the curve I'm creating new formulae adapted to the image in question. The mathematics of the formulae are invisible to me but the consequences of the formula are visible in real time in front of my eyes. Why would I do it any other way?  In Powerlord's image I would be able to look at those magnificent Ha jets (of which I'm very jealous! 😄) and shape the curve so that only the thin jet filaments made it above the red floor.

The Spanish Inquisition will accuse me of heresy or 'painting,' to which I can only say that PI users will do exactly what I do in the end and that is choose the final result based on subjective preference.

Cheating? Where are the rules?  We make our own. My rule is to try to make beautiful and informative pictures.

Olly

I do the same usually in Affinity Photo tbh Olly, this was an experiment in doing it diffferent ways - I've been trying to use Siril more and more now I'm getting thee hang of it a bit, using Sirilic sometimes, etc. It'll be about the 10th redo with this image!! But I'm going to start again again again (sic).🥵

I'll just do it the way I usually do in layers - I think by the time I got to that I was sick of looking at the thing and didn't have the patience. I shall set it aside for a week or so then take my time doing it right from the start - which will also get rid of those red/ha bleed stars. THEN I'll enter it into the compo. Have no fear @wimvb I have never won anything and doubt this'll be any different. 😞

stu

  • Like 1
Link to comment
Share on other sites

2 minutes ago, powerlord said:

I do the same usually in Affinity Photo tbh Olly, this was an experiment in doing it diffferent ways - I've been trying to use Siril more and more now I'm getting thee hang of it a bit, using Sirilic sometimes, etc. It'll be about the 10th redo with this image!! But I'm going to start again again again (sic).🥵

I'll just do it the way I usually do in layers - I think by the time I got to that I was sick of looking at the thing and didn't have the patience. I shall set it aside for a week or so then take my time doing it right from the start - which will also get rid of those red/ha bleed stars. THEN I'll enter it into the compo. Have no fear @wimvb I have never won anything and doubt this'll be any different. 😞

stu

Only ten?  I end up with files called M33 Final7_version12_starmod3_tuesday tweaksB1

Your image certainly has the potential to win a competition.

Olly

  • Haha 2
Link to comment
Share on other sites

57 minutes ago, ollypenrice said:

The Spanish Inquisition will accuse me of heresy or 'painting,' to which I can only say that PI users will do exactly what I do in the end and that is choose the final result based on subjective preference.

Cheating? Where are the rules?  We make our own. My rule is to try to make beautiful and informative pictures.

Interesting thing is that there is an accurate way of adding Ha data to RGB one - but no one is using it :D (and probably would not prefer the scientific result over layered composition or pixel math that is used).

 

Link to comment
Share on other sites

3 minutes ago, vlaiv said:

Interesting thing is that there is an accurate way of adding Ha data to RGB one - but no one is using it :D (and probably would not prefer the scientific result over layered composition or pixel math that is used).

 

Edited by ollypenrice
Can't type on this post!?!
Link to comment
Share on other sites

4 minutes ago, vlaiv said:

Interesting thing is that there is an accurate way of adding Ha data to RGB one - but no one is using it :D (and probably would not prefer the scientific result over layered composition or pixel math that is used).

 

I'd have thought the accurate way to capture Ha was through a red filter. In this case it is captured in its natural proportion, no?

If we add Ha we do so to add information in the form of faint signal and structural contrast. 

How are you definging 'accurate' here?

Olly

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
  • 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.