Colour balancing of data that was filtered by a light pollut

General discussion about StarTools.
Post Reply
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Colour balancing of data that was filtered by a light pollut

Post by almcl »

In the article about colour balancing, Ivo mentions:

“Correct colouring may be achieved by shooting deep luminance data with light pollution filter in place, while shooting colour data without filter in place, after which both are processed separately and finally combined.”

Does anyone have a workflow for this?

I have a stack of 5 min subs taken with a clip filter in place and another stack of (very much shorter) subs without the filter and am wondering how to make use of them? The layer module? Or LRGB? And what settings/modules?
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour balancing of data that was filtered by a light po

Post by admin »

You can even add both filtered and unfiltered data together (by creating a weighted synthetic luminance in the Layer module) to improve the luminance signal.

Simply process the luminance stack in b&w as normal until you're happy with the detail.
The process the color stack as normal until you're happy with that - the final image is allowed to be pushed more (and allows for you to be a bit more heavy handed with the noise reduction) than you would otherwise allow for.

My personal preference is to use the Layer module to merge color data of one stack with the luminance of the other by doing the following;

* Assuming you have the color data already loaded, launch the Layer module.
* Select Layer Mode "Color Extract fg" to extract the colors from the color image.
* Click Copy to store the output in the buffer.
* Open the luminance image
* Click Swap background and foreground (putting luminance in the background)
* Click Paste->Fg to paste the extracted colors into the foreground layer
* Select Layer Mode "Color Of fg" to use the luminance information from the background and the color of the foreground.
* Select Brightness Mask Mode "Where fg is dark use bg"
* Use Brightness Mask Power to subdue the coloring of the background.
* Use Blend Amount to increase/decrease saturation
* Use Filter Kernel Radius (with Gaussian filter) to mitigate color noise.

Hope this helps and do let me know how you get on!


The above gives you a lot of control over the final look and how the colors are blended.
Ivo Jager
StarTools creator and astronomy enthusiast
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Colour balancing of data that was filtered by a light po

Post by almcl »

Many thanks for that, Ivo.

Will have a play this coming (cloudy) weekend and see how it goes.
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Colour balancing of data that was filtered by a light po

Post by almcl »

Sorry it has taken so long to respond, it has taken me this long to get some decent data and even now have a monster dust bunny for which apologies.

I took a stack of 30 unfiltered light frames, stacked with flats and bias in DSS, processed in ST to have much more colour than would usually use.

I took a stack of light pollution filtered subs and processed those.

I then tried combining in the LRGB module using the filtered ones as a luminance channel and the unfiltered as the RGB.

Also tried the work flow above.

I am not sure I got any of this right, so comments and suggestions would be gratefully received.
Unfiltered stack
Unfiltered stack
unfiltered.jpg (208.12 KiB) Viewed 12443 times
Light pollution filtered stack
Light pollution filtered stack
Filtered .jpg (122.67 KiB) Viewed 12443 times
LRGB composite
LRGB composite
NewLRGBComposite.jpg (135.53 KiB) Viewed 12443 times
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Colour balancing of data that was filtered by a light po

Post by almcl »

And finally, the Layer module combined version
Layer module combination
Layer module combination
layered filtered and unfiltered.jpeg (176.68 KiB) Viewed 12443 times
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour balancing of data that was filtered by a light po

Post by admin »

Hi,

I'm noticing that your filtered and unfiltered images don't line up. Did you make sure to use one stack as a reference for the other when stacking?

The LRGB technique mentioned is certainly one way you could do it. There is another one which uses the Layer module. Much depends on your preference (I like the latter as it gives me more control).
Ivo Jager
StarTools creator and astronomy enthusiast
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Colour balancing of data that was filtered by a light po

Post by almcl »

Don't think I used the 'one-stack-to-align-another' technique.

I'll try and sort out a stack of unfiltered ones aligned on the filtered - but with nine different sets of data now, it's getting a bit complicated!
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour balancing of data that was filtered by a light po

Post by admin »

almcl wrote:Don't think I used the 'one-stack-to-align-another' technique.

I'll try and sort out a stack of unfiltered ones aligned on the filtered - but with nine different sets of data now, it's getting a bit complicated!
9 sets? :think:
Ivo Jager
StarTools creator and astronomy enthusiast
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: Colour balancing of data that was filtered by a light po

Post by almcl »

'Fraid so!

There was cloud, sensor marks, tracking failures, different ISOs, operator error, the dog ate my homework, the sun was in my eyes (!?)...

Now there's a dust bunny that the flats haven't dealt with, so data set no 10 is on the horizon!

In between times I did get a bit more dust detail:
Iris neb [8].jpg
Iris neb [8].jpg (271.93 KiB) Viewed 12399 times
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
Post Reply