Rosette Nebula Ha-OIII-RGB with an ED80 - help me settle this one

[graemlourens]

Hi everybody.

I can't seem to get to a point to be happy with this target. And probably i'm trying too hard and expecting too much, but before i go and finalize everything i'd love your opinions.

I have a three good masters, Ha, OIII and RGB, whereby i have chosen the path of first making an bi-color Image, and then pulled the star color of the RGB image to the bi-color image (separate thread discussing the ways to combine RGB & bi-color data here: 

I'm not 100% sure about having RGB stars in a bi-color image, but it the blue stars (or rather purple, as my scope squints!) give a nice contrast to the dominantly red ha, don't you think?
The thing bugging me the most is the red. As i've used mostly HA as LUM, the red is turning pink. I'm able to pull the saturated red back in PS, but i'm also not happy with that either...

I'm also not sure how far to go with Contrast enhancement. As the data is strong, i can go very far, but then it looks unnatural... trying to find the correct point to stop, so it does not look over-engineered.

I've looked at the image for too long, and i need some outside input!

All thoughts warmly welcome!

Kind regards, Graem

[Droogie 2001]

Very nice. I don't see much wrong with that, it has a real 3d feel to it.

 

 

[Adam J]

Nonsense I love RGB stars in a bi-colour image.

My only comment is that I would personally bring the OIII balance up a little more as its a little H-a dominant to my eye.

[graemlourens]

12 hours ago, Droogie 2001 said:

Very nice. I don't see much wrong with that, it has a real 3d feel to it.

 

 

Thank you for your kind words!

7 hours ago, Adam J said:

Nonsense I love RGB stars in a bi-colour image.

My only comment is that I would personally bring the OIII balance up a little more as its a little H-a dominant to my eye.

Adam you're right, i have already experimented with bringing the blues up a little (but leaving the green down), to get a more teal coloring. Though i wanted to keep the OIII kind of low key, as comparing to HA it is really not that strong, and i dislike the other images around there, that try to make the inner circle OIII dominated, as it is not the case if i'm reading my raws correctly.

General question as i'm not the best with seeing colors: is the red in the image the 'correct red' or have i drifted to far to pink?

Kind regards, Graem

[Adam J]

 

4 hours ago, graemlourens said:

Thank you for your kind words!

Adam you're right, i have already experimented with bringing the blues up a little (but leaving the green down), to get a more teal coloring. Though i wanted to keep the OIII kind of low key, as comparing to HA it is really not that strong, and i dislike the other images around there, that try to make the inner circle OIII dominated, as it is not the case if i'm reading my raws correctly.

General question as i'm not the best with seeing colors: is the red in the image the 'correct red' or have i drifted to far to pink?

Kind regards, Graem

To be honest its probably less pink than my attempt.. I would not say that I tried to artificially increase the OIII I just processed each image separately pushing them as far as the data would allow and then combined them.

I actually found the OIII signal in the centre of the Rosetta to be quite strong just less expansive than the H-a. But I am working in Narrowband with a DSLR so I doubt that I have the flattest response with wavelength in the world, if anything I would have thought that it would result in lower OIII signal for me but then again perhaps its a little because I did not push my H-a data hard enough.

You can see the low resolution version of my attempt in my profile picture. Yours is a better image mind you...just not sure I agree that the OIII is really that subtle. 

[Filroden]

5 hours ago, graemlourens said:

Though i wanted to keep the OIII kind of low key, as comparing to HA it is really not that strong, and i dislike the other images around there, that try to make the inner circle OIII

if you're trying to achieve a true balance I think you'd be better following Olly's technique of blending the narrowband into the RGB to enhance the structure in colour and luminance. 

I'm in the camp that finds RGB stars in a bicolour image 'odd'. I'd rather see the stars  as monochrome than in an alternate palette. However, this is just my subjective preference. I couldn't criticise your actual result which is good!

[graemlourens]

5 hours ago, Adam J said:

 

To be honest its probably less pink than my attempt.. I would not say that I tried to artificially increase the OIII I just processed each image separately pushing them as far as the data would allow and then combined them.

I actually found the OIII signal in the centre of the Rosetta to be quite strong just less expansive than the H-a. But I am working in Narrowband with a DSLR so I doubt that I have the flattest response with wavelength in the world, if anything I would have thought that it would result in lower OIII signal for me but then again perhaps its a little because I did not push my H-a data hard enough.

You can see the low resolution version of my attempt in my profile picture. Yours is a better image mind you...just not sure I agree that the OIII is really that subtle. 

Thank you Adam for your thoughts. I will definitively re-think this and play around more with the OIII. Its so hard to settle on it, when there are so many different versions out there, that are all right & wrong at the same time!

4 hours ago, Filroden said:

if you're trying to achieve a true balance I think you'd be better following Olly's technique of blending the narrowband into the RGB to enhance the structure in colour and luminance. 

I'm in the camp that finds RGB stars in a bicolour image 'odd'. I'd rather see the stars  as monochrome than in an alternate palette. However, this is just my subjective preference. I couldn't criticise your actual result which is good!

Hi Ken.

I indeed did play around with Ollys proposal, but the results are, except i'm missing something, extremely similar. As in this technique, pretty much everything except the stars get 'overwritten' by the OIII & HA you're adding into the channels, the only thing that is still original of the RGB image (after blending), are the stars, and as the stars in the narrowband subs are similar or even better (due to not having the blue bloat of the ED80). For this reason i did not see any harm doing it through the bicolor path Color wise i'm doing the exact same as olly's proposal, just instead of adding it into an existing RGB-image, i'm combining it into an 'empty' image (via tone mapping approach) so the balance of colors can be achieved in the same way, in both paths in my opinion.

Or am i missing something that will result in something different, coming from either path?

If there would be a larger are surrounding the nebula, that was visible in my fov, this would be a different thing, and i'd probably do ollys way (for example both Veil nebulas would be an example where i would not do it the way i did now, as the fov would be dominated by background & stars, and not nebulosity, so its more efficient to add 10% to 90% than the other way round).

Correct me if i'm saying something senseless

Concerning the preference of rgb stars in narrowband, it indeed is a question of taste, and i've not made up my mind yet!

Kind regards, Graem

[Filroden]

On 17/05/2017 at 15:27, graemlourens said:

Or am i missing something that will result in something different, coming from either path?

Here's my attempt at HaRGB

 

The luminance is largely Ha with some contribution from RGB, the colour is the reverse. I think it was a ratio of 33% Ha into RGB and 66% Ha into lum. I didn't get the stars right but the entire image shares the same panel. Had I an Oiii filter I think it would have served to enhance the central detail.

What did your RGB image look like?

[graemlourens]

13 hours ago, Filroden said:

Here's my attempt at HaRGB

The luminance is largely Ha with some contribution from RGB, the colour is the reverse. I think it was a ratio of 33% Ha into RGB and 66% Ha into lum. I didn't get the stars right but the entire image shares the same panel. Had I an Oiii filter I think it would have served to enhance the central detail.

What did your RGB image look like?

Hi Ken, i think we're missunderstanding each other. I was talking about HaOIIIRGB in both paths, being pretty much the same. HaRGB is obviously totally different  i'm not trying HaRGB at all for this target.

Update: Ken i think i know what you mean now (with your example), but i'm still not seeing any difference. I can achieve the exact same result in both ways, its just the order how i'm doing it that is different, so i still don't see why that way would be more 'realistic' - please somebody shed light i'm mistaken! In the end the 'realistic' feeling is an approximation (the luminosity is in any case not realistic relative to our eyes) whereby for the color there are obviously certain guidelines (OIII not being blue etc) that i'm really trying to uphold here.

My RGB-Image is pretty much flat background + stars (as i only did 5 min exposures unbinned) so there is not much contribution from the RGB-image except the star color. Maybe you did a much longer RGB-exposure? that would then be a different story if my RGB-image had already good signal of the rosette (what it really doesn't in my case)

Kind regards, Graem

[PatrickGilliland]

I like it - has nice structure to it and star colour looks good too

[Rodd]

Nice looking data.  I like the first one, as it reveals much more of the faint structure throughout, esp. the background.  However, I think it needs a slight downward tweak of the black point (not too much to clip)--but just a bit--I think that will really enhance the brighter areas.   But this is a quibble--its a glorious image.

Rodd

[graemlourens]

1 hour ago, PatrickGilliland said:

I like it - has nice structure to it and star colour looks good too

Thank you Paddy! Will still try and correct the purple stars (olly gave me a trick how to do this....) and maybe that will even improve it...

30 minutes ago, Rodd said:

Nice looking data.  I like the first one, as it reveals much more of the faint structure throughout, esp. the background.  However, I think it needs a slight downward tweak of the black point (not too much to clip)--but just a bit--I think that will really enhance the brighter areas.   But this is a quibble--its a glorious image.

Rodd

Only the first image in this thread is mine :)..........

But you're maybe spot on with the black point!! Now looking at it, i probably over estimated the background level. I'll definitively give this a try.... It does seem kind of 'washed out' and a black point clipping could solve that.

I'll keep you posted. First comes a weekend with my children (we've got exquisit weather here in warsaw...)

Kind regards, Graem

[Rodd]

20 minutes ago, graemlourens said:

Thank you Paddy! Will still try and correct the purple stars (olly gave me a trick how to do this....) and maybe that will even improve it...

Only the first image in this thread is mine :)..........

But you're maybe spot on with the black point!! Now looking at it, i probably over estimated the background level. I'll definitively give this a try.... It does seem kind of 'washed out' and a black point clipping could solve that.

I'll keep you posted. First comes a weekend with my children (we've got exquisit weather here in warsaw...)

Kind regards, Graem

Enjoy the little ones while you can--soon they will no longer be little.

[gorann]

I just think it is fabulous, soft and still sharp with just the right colours. Congratulations!

So, stop right there

[graemlourens]

9 hours ago, Rodd said:

Enjoy the little ones while you can--soon they will no longer be little.

True words  

7 hours ago, gorann said:

I just think it is fabulous, soft and still sharp with just the right colours. Congratulations!

So, stop right there

Thank you Göran! Difficult to find the line between soft & too contrasted. I'm happy you like it.

Kind regards, Graem

[gorann]

Maybe it is just me, but the colours and light make me think of some old master's oil painting - and I like that very much