Where is the blue color in my images?

[magnusl]

Hi!

I'm have a problem with colors in my DSO images - when I look at other peoples images of galaxies, there is often a lot of blue in them. However, I seem not to be able to produce that blue color. Anyone can guide me here, or help understand why?

I've uploaded an xisf-file at:

https://www.dropbox.com/sh/fs744ujp8oed79d/AACfnztnt-E98N8km8Tv2dUBa?dl=0

This one is stacked, 87 subs of 300secs, taken with a C8 f:6.3, DSLR Canon 350D astro modified, with a CLS filter attached. The image is postprocessed up until it is time for saturation, and that is where I fail. When increasing saturation, I get orange and yellow, some red, but little blue.

Any ideas warmly welcome!

Magnus

[msmythers]

Magnus

The blue is weak so it must have a very heavy saturation stretch. I normally do some of the color saturation before actually stretching the image to the non-linear stage but that's me. I also don't use a CLS filter and have no idea if that might be contributing to your issue.

As far as your image hear is what I did. I made a star mask and then a range mask of the galaxy. I lowered the upper limit of the range mask to protect the core of the galaxy.  There seems to be a very odd purple in the core. I subtracted the star mask from the range mask with PixelMath. I applied the range mask to your image. I then used the ColorSaturation tool with a range of 10 to maximize the saturation. The curve I used is also to maximize the blue.


Mike

[Niall Saunders]

I assume that you are using the ColorSaturation process - but you might be better off starting with Curves, and putting maximum effort into the individual colour planes, one at a time, whilst also monitorig results in the RealTime Preview window (also set to monitor only the colour plane that you are currently trying to enhance).

You can also then play with the Saturation channel in Curves - this time monitoring RGB/K in the RT window.

Just remember - Curves is a process that really wants to have 'small' adjustments, perhaps with a few iterations, rather than one 'big' adjustment.

If I get a chance, I will download your image and have a play, but others may have more time than me.

[magnusl]

Hi!

Great thanks!! Yes, I've mostly tried curves this far. But it's good to hear that you too find that there is little blue and a strange purple.... so it's not just my lack of understanding PI. But I wonder why these strange color effects are there. The CLS should at least not affect the blue, as I understand it. So any suggestions on the reasons for this is also warmly welcome.

And I'll play along with your suggestions in the meantime!!

Best,

Magnus

[Niall Saunders]

Hi Magnus,

I find that I am usually missing virtually all colour from my OSC images until quite late on in the post-processing phase - typically, I only see any colours start to appear as I apply Curves to the various channels (R, G, B, L and S) as I described earlier.

To begin with, I was always concerned that it must be as a result of something that I was 'doing wrong' during image acquisition, or in pre-processing. Now, I just don't get too worried - unless, of course, I don't get any colour during the Curves stage of post-processing !!

Take your time - patience is the key.