Colorizing California

[Galaxyfaraway]

Hi,

I have tried narrowband for the first time the other night but unfortunately I only have the HA filter as the other two are still on backorder.

While waiting for the other filters to arrive to make a (proper) colour image, I wanted to try and colorize this one for fun. I was wondering if there are any tutorials out there on how to do this in photoshop? I have seen some people use bicolour techniques and some paint in colours into the data of even just one channel (creating extra channels synthetically) but I can't quite figure out the tools to use to achieve this. I realize this is scientifically inaccurate and has little merit but it is just to pass time till I get my other filters (and to improve PS skills).

Thanks for the help!  gfa

ps: these are 5 subs at 20mins guided exposures, taken from London (red zone). Almost no processing. WO GT81 and Atik 490EX and Astrodon HA 5nm filters were used.

[cfpendock]

Noels Actions for photoshop has a procedure which results in this from your image (quality down because of JPEG).

Chris

 

[Galaxyfaraway]

Interesting. But just the red hue? Is there not an option to assign different colour values to different nebula textures and also brightness? Just curious.

Perhaps I wait for the filters. This needs a lot of work. Also to get the RGB stars in.

Thanks, gfa

[ollypenrice]

That's a great capture, sharp and contrasty. NB is a lifesaver from heavy LP.

One way you can colourize is simply by saving the image in RGB mode in Photoshop (Image-Mode-RGB.) You can then manipulate the Curve for each channel separately. To drive the image towards red you could pull down the blue channel heavily, even black clipping it hard, then lift the red and balance the result with the green Curve. Other tools at your disposal are the colour balance sliders (which just manipulate the Curve through a simplified interface) and the Selective Colour palettes. In order for the image not to become just a monochrome of a different colour you need to give the different channels curves of different shapes, so making the colour balance at low brightnesses different from those at high brightnesses. Although I haven't done this, I guess you could re-apply the orginal as a luminance layer to resotore the original brightnesses while retaining the colour.

I see others have just replied so I'll see what they say. This isn't really my area of experience!

Olly

[johnrt]

Very nice, I always think the nicest images of the "Californee" make it look like a big fluffy pillow, and your mono certainly gives it that quality!

[steppenwolf]

Here's a down and dirty quick process for you using completely false and arbitrarily placed colours with zero scientific contribution - but you might think is looks pretty!

I produced this using layers and layer masks to colour certain features, the colour being developed using the Hue/Saturation tool.

 

 

[ollypenrice]

Heheh, that's great, Steve. You're an Architect of Nature, my boy...

[Galaxyfaraway]

Yes! 'Down and dirty' as well as 'scientifically inaccurate' might have to do till filters come 

i was just curious how people achieve this effect: using techniques with selection tools typically used to colorise for example b&w portraits seems awkward...

is there an official name for this technique? (The 'Messing around' actions tool perhaps). It might be possible, though vaguely, to try and reproduce similar colours this way, obtained normally using 3 filters, by comparing it to an actual Hubble-palette image and paint in. But some of the nebulae details will always be missing plus I suspect this is something not very acceptable/accepted.)

Edited February 5, 2016 by Galaxyfaraway

[johnrt]

1 minute ago, Galaxyfaraway said:

Yes! 'Down and dirty' as well as 'scientifically inaccurate' might have to do till filters come 

i was just curious how people achieve this effect: using techniques with selection tools typically used to colorise for example b&w portraits seems awkward...

is there an official name for this (lack of) 'technique'? (The 'Messing around' actions tool perhaps). It might be possible, though vaguely, to try and reproduce similar colours this way, obtained normally using 3 filters, by comparing it to an actual Hubble-palette image and paint in. But some of the nebulae details will always be missing plus I suspect this is something not very acceptable/accepted.)

Sara @swag72 has just raised this potentially divisive topic herself with a (very nice) 1 filter rosette nebula colourised to the HST colours. She could tell you more about the overall reaction she got to it in terms of it being seen as acceptable or not.

[steppenwolf]

Quote

is there an official name for this (lack of) 'technique'? (The 'Messing around' actions tool perhaps).

Sara (SWAG72) is the person to ask about this as she really has got this technique sorted although I know she has only used it as a 'test piece' to prove a very interesting artistic point.

[Galaxyfaraway]

Fascinating. Thanks for this. And for the demo of image manipulation. Much appreciated. Gfa

[Dave In Vermont]

Here's it back after I downloaded a copy. Now it's a PNG instead of a JPEG.

Dave

[swag72]

There were a couple of interesting debates here and here - I coined the name 'Monochrome Colour Mapping' as this was in effect exactly what I was doing. 

It created a pleasing and unscientific image - Some felt this was OK as long as it was fully disclosed, others disagreed and felt it had no basis......... it was an interesting debate

[Galaxyfaraway]

Thanks Sara - that's very interesting. I have to read up on the whole human colour perception & interpretation to really appreciate why assigning certain data from separate narrowband filters to different channels and combining those, is more 'scientific' than colouring in into images. I sort of get it (different wavelength etc) but since the colours can be interchangeable in terms of assigning them to different RGB channels, I don't think I get it 100%. I presume it's not the colour that matters but the fact that only certain parts of the nebula will show up on an image under a specific wavelength and a colour channel is there to specifically highlight these parts. 

Sorry if I got this completely wrong.