Hi all,
I'd like to ask for some feedback on my Pelican Nebula: https://www.astrobin.com/sbf9nw/C/
I had quite a lot of problems with this data (it was very humid) so I am happy to at least get something out. I also like the framing, which was one of first projects for me that I intentionally frame over multiple nights.
I think, I had some problems with flat calibration which usually works for me flawlessly. Blue channel (Oiii) was all around the place so it make the background extraction really hard, so I did it manually with color masks (based on combined Ha + Sii). So I know it's not perfect, but at the end I was quite happy with that. I also had huge issues to get some usable stars (I know the Cyg 56 has a lot of strange patterns, I could maybe clean it up a little bit better).
I used 'my standard' PixInsight flow (roughly):
- Background extraction - Usually I'd do it ASAP, but I actually couldn't do it with DBE/ABE so I 'equalize' it later in the processing, I even tried it with Graxpert without luck
- Combine SHO & RGB channels
- StarXTerminator (throw away SHO stars & RGB starless)
SHO starless:
- BlurXTerminator
- Stretch to linear - with STF + histogram transformation, I intentionally didn't stretch a lot, I prefer to do it later with Curves
- ColorMask with original (starless) Ha+Sii (I use PixelMath to create mask) - lowering the background, trying to keep the lighter parts unchanges as much as possible
- ColorMask with Magenta, Green & Cyan (to remove magenta color, do the blue-orange)
- ColorMask green - to remove very green trend towards the left side (should usually be removed with background extraction)
- Final curves (on full image and with color masks)
RGB stars:
- BlurXterminator (very easy processing but with 'correct first' option)
- SPCC
- Stretch to linear - manually with histogram transformation
Combine SHO starless + RGB stars
- light NoiseXTerminator for finalization
Hopefully I didn't forget anything.
I'd like to ask for some feedback on my Pelican Nebula: https://www.astrobin.com/sbf9nw/C/
I had quite a lot of problems with this data (it was very humid) so I am happy to at least get something out. I also like the framing, which was one of first projects for me that I intentionally frame over multiple nights.
I think, I had some problems with flat calibration which usually works for me flawlessly. Blue channel (Oiii) was all around the place so it make the background extraction really hard, so I did it manually with color masks (based on combined Ha + Sii). So I know it's not perfect, but at the end I was quite happy with that. I also had huge issues to get some usable stars (I know the Cyg 56 has a lot of strange patterns, I could maybe clean it up a little bit better).
I used 'my standard' PixInsight flow (roughly):
- Background extraction - Usually I'd do it ASAP, but I actually couldn't do it with DBE/ABE so I 'equalize' it later in the processing, I even tried it with Graxpert without luck
- Combine SHO & RGB channels
- StarXTerminator (throw away SHO stars & RGB starless)
SHO starless:
- BlurXTerminator
- Stretch to linear - with STF + histogram transformation, I intentionally didn't stretch a lot, I prefer to do it later with Curves
- ColorMask with original (starless) Ha+Sii (I use PixelMath to create mask) - lowering the background, trying to keep the lighter parts unchanges as much as possible
- ColorMask with Magenta, Green & Cyan (to remove magenta color, do the blue-orange)
- ColorMask green - to remove very green trend towards the left side (should usually be removed with background extraction)
- Final curves (on full image and with color masks)
RGB stars:
- BlurXterminator (very easy processing but with 'correct first' option)
- SPCC
- Stretch to linear - manually with histogram transformation
Combine SHO starless + RGB stars
- light NoiseXTerminator for finalization
Hopefully I didn't forget anything.
I am fairly happy with my Pelican Nebula but something I think is missing... and it was though data IMHO to work with 
). I used Ha+Sii as a mask for BG neutralization but probably if I extract L on final image and just try to suppres BG a little bit more, it could give great result!