9 July 2020 - APP 1.082 has been released which contains one important bug fix. 1.082 has full Fujifilm RAF support, so that includes SuperCCD & X-Trans camera's 🙂 !
9 July 2020 - New and updated video tutorial using APP 1.081: Complete LRGB Tutorial of NGC292, The Small Magellanic Cloud by Christian Sasse (iTelescope.net) and Mabula Haverkamp
2019 September: Astro Pixel Processor and iTelescope.net celebrate a new Partnership!
Star Color Calibration Issue
The mean differences between the channels R,G,B are from 2.53 to 3.02, too low. So RGB il almost white. The mean difference between Ha and the other channels is ~ 11.12, 4 times more.
The rgb data is split channels from Broadband filter With osc camera zwo 294 and the filter is L pro. The Ha data is extract from L enhance narrowband filter ... this is the workflow that I following https://www.astropixelprocessor.com/community/tutorials-workflows/combining-r-g-b-with-ha-oiii-data-using-an-optolong-l-enhance-filter/
Yes that could be one of the issues. You can always lower the amount of Ha you use for instance. Mixing it 50-50.
I'll have a look today.
So here's what I got. I selected stars on the left hand side and another big box underneath the galaxy. It seems that the filters do change the broadband data a bit, so I tweaked the slopes a little bit to have the line go through the population a bit better (that's the white line in the correction graphs). I also made a comparison with 50% Ha into the red and 33% Ha into the red. Before I combined all panels, I first did light pollution correction to have a nice background, all similar to each other.
This is right after combination with 50% Ha;
This is with color correction and a tweaked slope of the line. I also lowered both Magenta - Green slider to about -0.05 to remove a bit of green cast;
Next up, 33% Ha into red;
And with tweaked slopes;
Saturation was upped a bit to 0.19. Is this more to your liking? It does seem the colors could be more pronounced, what was the total integration time?
Hello Vincent ...
For sure is better than my processing ...
the total time is about 10 hours ... 7 rgb and 3 Ha
I think that before star color calibration you still get better result ...
What you think ?
No, personally I don't think so. That first image looks very dull, color wise. So there seems to be an offset, created probably by the filters.
I can’t have good view from screenshots ... plz when you finish upload it in server to check it ...
Ofcourse, I uploaded the 33% Ha into Red final stretch in your directory on the server. The file is called "combine-RGB-image-33Ha-cbg-St.fits"
So to recap;
- Loaded all integrations and performed light pollution correction on each of them.
- Combined them in HaRGB, 33% Ha into red.
- Did another light pollution correction.
- Color calibration, large boxes around the galaxy and tweaking of the slopes to better match the star population.
- I removed some green bij moving the sliders (both of them) from green towards magenta with a value of -0.05.
- I added saturation, Sat. 0.19.
- Saved the stretch and uploaded it to the server for you.
Just so you have the details of this stretch. 😉
Thats great Vincent...
really I appreciate it ...
I will continue in hsl step ...
clear skies ...
I try to process my data from m51.
My files are R G B and HA from OSC
Everything going normal before Star Color Calibration after this my data taking a reddish color and i cant get over it ...
This is Before star Color
And this is after
Any Help will be appreciate it ...
If someone want i can upload my data to check it ...
Thank you Stratos for sharing your data with us 😉 I have had a go myself since I am actually working to improve HaRGB and LHaRGB composites for a future APP release.
The workflow to do star color calibration on a HaRGB (or a LHaRGB) is not recommended, in fact it is physically a wrong step. You need to do star color calibration on broadband data. If you add narrowband data in the mix and then try to perform star color calibration on a broadband+narrowband composite, it would be mere coincidence if the colors look good...
Star color calibration needs broadband data to make sense of the star color physics. Adding narrowband data in there will cause the R/G B/G ratio's of the stars in your data to be polluted by the narrowband emissions. Those narroband emissions don't follow the physics at all that we use for star color calibration.
So how to do this properly then?
1) create a RGB (or LRGB compositie), in this case I used formula RGB 2 in the RGB Combine tool
2) perform Remove Light Pollution and Background Calibration on the RGB composite ( ! if you perform Remove Light Pollution, you automatically perform Background Calibration, there is no need to do it again after Remove Light Pollution)
3) Perform Star Color Calibration on this RGB composite that is corrected for light pollution and is background calibrated:
4) We know open the RGB Combine tool and will add the H-alpha layer. Now, this next 2 steps are essential to preserve the RGB Star Color Calibrated colors in your composite:
- set normalize to none in the RGB Combine tool.
- Load the RGB star color calibrated result into the tool with the RGB 1 formula
These 2 steps will preserve the colors of the star color calibration 🙂 !
5) Add the H-alpha layer in the RGB Combine tool.
6) To preserve the colors and luminance of your star color calibrated result, set the luminance for the composite so that it is preserved. Normally you want Red to contribute for roughly 20%, Green for 70% and Blue for 10%.
7) And slightly add H-alpha to red, I added it for 23% to Red as can be seen in the below screenshot. Now to maintain the star colors, you need to gently adjust the multipliers for the R,G,B channels as can be seen as well. Red 1,15, Green 1,17, Blue 1,20. These factors do depend on the data, you need to do this carefully. Try to get the same colors as the star color calibrated result off course 😉
I am working to make this more automatic and easier for the user off course, but this is a proper workflow in which you can get good star colors with narrowband included to enhance the active regions in this case of M51 😉
This show the star color calibration result next to the final HaRGB composite: we see that star colors are still good and we also see much more enhanced active regions in M51 from the H-alpha data, top is the RGB composite after star color calibration, below is the finals HaRGB composite with the same star colors 🙂 :
Let me know if this is clear and/or if you have any questions about this workflow 😉
Crop of the field of view of the final HaRGB:
Owkee, well, I'm learning something as well there. 😉 I was assuming that, since there is broadband data, it would still work. But makes sense indeed, that's why I needed to fiddle around with the slopes apparently.
First of all I must thank you for this great explained workflow and saw us where was our mistake ...
This is great because we know now to process narrowband and broadband data ...
The last years color cameras with narrowband and broadband filters are so popular and seems like the came to stay ...
This will be great tutorial for all people that use this kind of equipment...
I will try it this tutorial and I let you know how went ...
ps you can upload your version in server that can I take it to compare with mine ?
Great Mabula, perfect !
Μάστορας ο Mabula !
we learn a great tutorial here ...
Πολύ καλή επεξεργασία ... είναι μάστορας τέλος ...
You can download my results here for comparison:
Thank you so much Mabula !!!
really I appreciate it...
This workflow will be so helpful for all
Narrowband and broadband users ...