[RCC] M51 first time using H-alpha (data linked)

I'm not sure who that guy is, but he literally tells you to mask 90% of the photo and then add in the Ha.      You shouldn't mask when you add data.  Might as well just not use Ha if you're going to do that.   I stopped watching after that.

Try this:
https://www.nightphotons.com/guides/advanced-narrowband-combination

You should consider continuum subtraction for a much more accurate image.    Also, as expected it looks like the Ha falls off rather quickly in your image, with nothing in the dimmer parts of the spiral.  Maybe change the color of your Ha a bit, it feels pink.


Anyways, try the guide, no masking required.  I've been using this for over a year now.

Thank you, I will try those methods.

Johannes

Arun H:
I second the use of the method linked by Brian Puhl. 

You will need to remove the stars before doing the H-alpha addition to avoid artifacts in the stars. At least I have found that to be the case, although Hagen claims it is not needed.

StarX and Starnet will remove smaller Ha structures like in galaxies.  Highly recommend not doing that.   The any artifacting in the stars will 95% of the time get covered with your broadband stars.

This is not a complete image, only a WIP, but this is like 3 hours of RGB with my 36+ hours of Ha. (I just wanted to see the structures) Absolutely no StarX or Starnet was performed in the Ha layer.   

StarX would have eaten all the little structures in the bands of M106

Arun H:
I ended up getting saturated cores that stood out if I didn't remove broadband stars. What am I doing wrong?

Did you remove the stars from the broadband image before adding the continuum subtracted Ha?

The broadband stars go on top of the narrowband stars

Arun H:

Brian Puhl:
Did you remove the stars from the broadband image before adding the continuum subtracted Ha?

Yes. Essentially I had to combine two starless images and then add back the broadband stars. If I didn't remove the broadband stars prior to addition, I would get saturated cores that stood out.

Hmm. I think I'm (were) just getting confused here.   

You do need to remove broadband stars, at least as far as I know.  You don't need to remove narrowband stars.   

That being said, for bright targets that don't have small structures (like those found in galaxies) I will do starless prior to continuum subtraction.    It's all about preserving the details. StarX can be destructive if not used carefully.   Bright nebula and such you don't risk losing data. 


My generic flow:

Create starless RGB image
Combine HRR to perform continuum subtraction.  
Curve, denoise, and (almost) clip NB signal.  
Use pixelmath to re-add Ha NB.   Perform multiple times if needed to boost the signal/structure. 
Add RGB stars on top.   

That's exactly what took place in the example above.   The RGB data is weak in that tho, so don't read deep into it.   It was purely what I had that I could share.