Colour can be a subjective thing at times and some of you have expressed some hesitation or are a bit iffy about the colours I have arrived at in this image, so I am interested in your opinions here, which of the following three representations of the Swan do you think is the most accurate or best and why..?
I'm a DSLR girl, so I'm partial to the kinds of colours they produce.
But I believe that every image is unique. Everyone has their own interpretation they give their data. If we all processed every image the same, what would be the point? We may as well shoot each object just once if that were the case.
Colour can be a subjective thing at times and some of you have expressed some hesitation or are a bit iffy about the colours I have arrived at in this image, so I am interested in your opinions here, which of the following three representations of the Swan do you think is the most accurate or best and why..?
Mike
David Malin's AAO shot Mike. Because it was done by his understanding of object colors. He wrote a whole book on it. I think in the end, your image is over-saturated, over colorized. That's not to say it's bad, nor are you the first one to do this. It's just very easy to push an image too far with that much data. And... you have some excellent data there to work with. Take a look at Martin Pugh's approach to color. He has found the right balance of contrast and saturation.
If you keep this up, we will have to take that 12-inch away from you. I could do some nice H-a work with it from Coona.
I think RJ's got it. The AAO image is very similar but I think the advances in CCD over film allowed him to capture a few nuances there.
This neb is not straight pinkish red it has some subtle colours in it including white areas on the spine where the O111 mainly is. There is some brown dusty areas too.
My personal opinion on Ha O111 is less is more as there is so much Ha around it can easily take over an image and you lose the other subtle colours that can make an image more interesting.
Its not a matter of right or wrong though as it is one way of showing the object. You didn't invent those colours you simply accentuated some over others. So its a choice of display of the object. The standard if there were one would be what would this object look like viewed through human eyes if they were able to see it fully illuminated. As human eyes can't then this makes it complicated.
I'll vote for RJ too, the colours in that rendition look very natural and detailed. These nebulae most often display a whole range of subtle hues which are a delight to see. One exception that comes to mind is the Lobster which I'm working on at the moment, it is just red, very red!
I have no narrowband experence at all, but purely guessing I assume maybe Ha can easily become overpowering when combined with the rest of the data. Depends on what detail you are after though, if you want to show the full extent of Ha details well then that's probably also the colour you'll end up with - which is only natural
Amazing shots Mike - as always, I should add.
There are no rules for colour unless you specify what you are trying to accomplish with your image. And even then it is a murky subject.
If you're going for 'as the human eye sees it', then you are limiting yourself quite a bit. The equivalent in the luminance space of such a limited (and dare I say 'old-school') approach would mean no dynamic range management at all, washed out details, etc; not much to see really. It's all we could do before CCDs came along.
Simple (L)RGB imaging makes it hard to discern detail - all the interesting stuff such as Ha, SII sits in the same band (red), while the other interesting stuff is way overpowering (Hb, OIII) in the green band due to the way the human eye works (much more sensitive to green). Bringing out red more (for example by introducing higher saturation) to compensate for the human eye's comparatively poor response, can upset your colour balance and/or amount of discernible detail and so the problems really start if you're trying to stay 'true' to 'how the eye sees it'.
Today we have the unique ability to emphasise and bring out detail with colour as well as luminance, so why not use it? Personally, I am all for taking some liberty with colours, if it helps understanding an object better or if it helps draw the attention of the viewer to what you're trying to show. The HST palette is one such method to accomplish that goal for narrow band imaging, but why wouldn't you modify colours in RGB images to accomplish the same goal?
I'm a big fan of Mike's work in that he is absolutely not afraid of experimenting with colour (though his luminance data isn't too shabby either ). Old rules are there to be broken, especially if you have a good reason. I can think of plenty.
the AAO image looks noisy and the color shades depend on the noise, too. don't they. that's where the yellow tint (the brown) comes from.
the 2 images that have a clear black (as in 'no noise') as background http://www.cosmotography.com/images/small_m17.html
and http://www.astro-pics.com/17plm.htm
are less brown and overall show more differences in the nebula color.
myself being far from producing such images, I still have my own taste - and that goes for more nuances.
the pink in your image, michael, is stunning and like candy floss.
a looker.
but not my cup of tea.
Color can be so subjective, the shots I've taken with a OSC camera turned out just like the Anglo version, galabanys version is pretty but lacks the depth, I'm yet to be convinced the white in his is a true representation.
If you like it so be it, surely thats all that matters....... Unless you enter a competition, then if in doubt just use narrowband colors You can't go too wrong there.
On a personal level bit less pinky would do me, but I'm not saying thats correct either.
No real consensus yet about the three images other than AAO and RJ got a couple of votes and I Got none due to being a bit pink...what's wrong with pink anyway as Silv said like Fairy Floss..mmmmm fairy floss...mmm
Nice image MIke.
I really like to see the blues in the "spine" of M17 showing through, with pink around. Peter W posted a wonderful version some time ago.
Still, this has NB mixed in, and its.....well.....humungous really
Thanks for the comments Al, Ross and Rob, I'm leaving it as is....for now
Funny though, there has been over 500 IIS viewers of this thread now but only a fraction of that number have had a look at the full frame image...interesting Is this a case of if you post a thumbnail in the thread then most will just look at that..? cause that was only a low res crop
Seeing this is your image, its done to your taste, and thats what is great about this hobby in that we all can be a little creative with the end result.
Seeing this is your image, its done to your taste, and thats what is great about this hobby in that we all can be a little creative with the end result.
Its certainly a dazzling FOV, very nicely done.
Cheers,
Steve
Thanks Steve and you hit it on the head - Dazzling is exactly what I had in mind
Star are a little soft (well... for an A-list image) ...but I'm thinking high-res astrographas have *exactly* the same seeing problems as similar aperture long FL scopes.
Last edited by Peter Ward; 21-07-2012 at 09:01 PM.
Reason: typos... & I hate ipad keyboards
Spent a bit of time calibrating the colour in this one....
Yeh?...against what..? Doesn't look much like RJ's. Good to see that actually, my higher res version using the best subs.. stacks up quite well ...considering my crap conditions
Quote:
Nice image Mike
.
I'll take nice, nice is..?.. good
Quote:
Star are a little soft (well... for an A-list image) ...but I'm thinking high-res astrographas have *exactly* the same seeing problems as similar aperture long FL scopes.
Yep, with 1.57"/pix one needs better seeing to realise the potential
Last edited by strongmanmike; 21-07-2012 at 09:51 PM.