This is my second dataset with the DeltaRho.  I'm continuing to run it unguided, which is OK, though I'm definitely losing resolution.  The masters are above 2" FWHM.  But this data does look sharp.  Two reasons for that:  1) The wider the FOV the sharper things appear.  So by doing a mosaic I get more of a feeling of sharpness.  2) This data has a pretty good SNR and that allows BlurX to work harder.  Please do use full resolution and zoom around the nebula.  

The DR still has astigmatism.  BlurX did a good job on cleaning it up but some of the stars still display some artifacts.  I'm hopeful that we can make further progress on the astigmatism soon.

The data here are SHO with RGB stars.  Initially I created a joined version of the SHO and another of the RGB stars, but each master had slightly different star locations.  This would mean that the RGB stars would not match up to the SHO data.  After thinking about it for a bit, I decided to try replacing the SHO stars with RGB stars in a linear state before I joined the panels.  This would allow me to join everything at the same time but break them apart later for separate processing.  So my process steps were something like this:

1) BlurX correct only on each of the six channel masters
2) Create a straight SHO master
3) Run StarX to remove the SHO stars
4) Create an RGB master
5) Run StarX to separate the RGB stars
6) Add the RGB stars to the starless SHO master using this pixelmath formula in each Channel: mtf(.005,mtf(.995,SHO)+mtf(.995,Stars))
7) Join the two panels with Star Alignment in Mosaic Union mode
8) Run DeepSNR on the resulting SHO+RGBstars mosaic master
9) Run StarX to separate the RGB stars from the SHO master
10) Run SPCC on the RGB stars only master (I didn't want to alter the SHO data with this color calibration)
11) Run BlurX on the SHO master using a manual PSF
12) Separate the SHO master into S, H, and masters using Channel Extraction process

Now I could process the joined S, H, and O masters as I would any other narrowband image and add back the RGB stars at the end.  I chose to use a Foraxx narrowband combination.  This uses the OIII channel as a factor and blends the S and H into Red and Green with a mix that is determined by the strength of the O in that pixel.  I do that manually rather the use the Foraxx script.

Most any SHO combination pushes the H data way down and brings the S and O way up.  That costs a lot of SNR because the H has the highest SNR by far.  To resolve that SNR loss I created a Psuedo-Luminance.  In this case I simply combined the S, H and O masters in equal weights in Pixelmath.  (S+H+O)/3 = Pseudo Luminance.

I used approximately the same stretch on the Pseudo Luminance as I did on the SHO.  Then I could add the Pseudo Lum to the SHO as Luminance.  That really helped the SNR in the fainter areas.

At this point I moved the Starless SHO to Photoshop for finishing touches.  Once I was done in Photoshop I opened the 16 bit tiff version in Pixinsight and (after stretching the RGB stars to taste) I added the RGB stars to the SHO tiff using the same Pixelmath expression I mentioned above.

These three globes that seem to be attached to the primary nebula are my favorite part of the object.  Each of these could be an image of its own with a long FL scope.  

I noticed a couple of planetary nebulae in this FOV:

This tiny ring measures about 15 pixels wide or about 12 arc seconds.  I found it on Simbad as PN G295.6-01.5.  You can find it on the left edge about 20% of the way up from the bottom.


Also the well known PK 294-001 is in there and was picked up by the annotation.


I haven't done a mosaic in quite a while so it was nice to get this to work!