Back to forums

Red channel overpowering in M33 - color balancing advice needed

Processing Color Calibration Stretching RGB Deep Sky Galaxy M33
7 replies•190 views
Michael Telesco avatar
Michael Telesco avatar

Hi everyone,

I recently finished collecting data on M33 and had some questions about galaxy processing. I’ve noticed this before in other images however since I rarely image galaxies, I haven’t found a perfect solution.

Oftentimes in my final stack, the Red channel has a much stronger overall signal. This leads to issues with color balancing even with linear fit and SPCC/PCC. I have seen wonderful images of M33 with a beautifiul yellow core and outer blue/red ish arms, yet I can’t seem to recreate that with my current workflow. Does anyone have advice on how I can better incorperate my Red channel to not overpower the R/G? Or any go-to stretching techniques that may alleviate this? I am also curious if this is a common issue with bright galaxies in RGB.

Thank you so much!

CS,

Michael

Well Written Respectful Engaging
andrea tasselli avatar
andrea tasselli avatar
What camera/what filter(s)/what scope?
Willem Jan Drijfhout avatar
Willem Jan Drijfhout avatar
In SPCC you can choose your white reference. Default this is set to 'Average Spiral Galaxy', but you can choose from a whole list. If you choose S0 Galaxy for example (pretty far down), the resulting image comes out significantly cooler, with less red and more blue. You may want to experiment with these values. You can compare this setting with the Whitebalance setting in your photo camera, where 'sunny' or 'cloudy' give quite different overall looks.
Alternatively you can always use the nuclear option: after initial stretch, make an additional stretch (probably unstretch) on only the red channel in HistogramTransformation, and make sure the red peak in the histogram overlaps the green and blue ones.
Michael Telesco avatar
Michael Telesco avatar

Takahashi Epsilon E160ED, Astronomik Deep Sky RGB.

The issue isn’t just that there is an overcast of Red but moreso there are Red parts that overpower significantly other parts of the galaxy. If I stretch the Red channel less, it doesn’t quite fix the issue.

I saw a recent post here about an M101 process and it looks sort of similar to what I am getting with a red bias in parts of the galaxy where there shouldn’t be.

andrea tasselli avatar
andrea tasselli avatar
You might have a faulty filter (e.g. passing near IR) but that is quite rare nowadays. Can you post the raw files so that we can assess what might be? Post meaning can be downloaded.
Scott Badger avatar
Scott Badger avatar

Not sure if it has anything to do with your issue, but you shouldn’t use linear fit on the channels before combining. At best, you’ll remove it with calibration anyhow.

Cheers,

Scott

Arun H avatar
Arun H avatar
Scott Badger:
Not sure if it has anything to do with your issue, but you shouldn’t use linear fit on the channels before combining. At best, you’ll remove it with calibration anyhow.

Cheers,

Scott

I agree that linear fit should not be necessary if you are doing something like SPCC. However, I don't see the actual harm in it. It simply adds or subtracts a bias and multiplies everything by a linear factor. Photometric calibration computes the relative weighting of the channels and these factors should not in the end matter one way or another as you have mentioned.
Well Written Insightful Respectful Concise Supportive
Michael Telesco avatar
Michael Telesco avatar

Huh just found the issue.

It seemed my gradient correction incorrectly shifted my Red signal which gave it a bias. I just did an RGB with no gradient correction and it fixed the issue. Guess I have to be more careful with that!

Thank you guys again.

CS,

Michael

Concise