Bubble Nebula

User images created with StarTools.
Post Reply
MrMxyzptlk
Posts: 23
Joined: Tue Oct 14, 2014 9:46 am

Bubble Nebula

Post by MrMxyzptlk »

Hello to you all. I'm new to Star Tools, I'm pretty new to astrophotography as well as I only began deep sky imaging last December. The last few nights I acquired data for M52 and the Bubble Nebula, I have stacked a total of 4 hours worth of data (I had more, but I threw out the bad data), calibrated with flats, darks, dark flats and bias frames.
Image
You can see the full result here:
http://www.astrobin.com/127890/

I decided to ask for some help, as I think that there might be more data in there or better processing options, I used mostly the M8 tutorial to work with this acquisition.
Here is the link for the raw stack, the stack is not white balanced, it has stacking artifacts on the margins, it was done with IRIS (raw decoding, calibration, cfa2pic, co-register, norm, composite)

https://www.dropbox.com/s/nqx9zgk4oj4ky ... e.fit?dl=0

I must note that my Baader BCF filter in the Canon 1100D is not aligned perfectly whit my sensor, so I have some artifacts/glare at the bright stars, I'm curious how this can be eliminated.

So if you have some time, maybe you can show me some better processing techniques. Thank you in advance
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Bubble Nebula

Post by admin »

Wow... :shock:
That's an absolutely stunning debut! Looking at the astrobin image, it's very, very hard to fault for the equipment used; quite an accomplishment for someone who says he just started! :bow-yellow:
It's getting late here, but I'll definitely have a play with the data tomorrow.
Ivo Jager
StarTools creator and astronomy enthusiast
MrMxyzptlk
Posts: 23
Joined: Tue Oct 14, 2014 9:46 am

Re: Bubble Nebula

Post by MrMxyzptlk »

Thanks. I don't have any complains about my equipment beside the BCF filter in my Canon. My only complains are my suburban sky, with some light pollution :)
I'm looking forward to see your processing and learn. Good night :)
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Bubble Nebula

Post by admin »

Ok, I had a go at it. Lovely data! You really got your acquisition down pat - something a lot of people still struggle with, even years later!
The data clearly has not been pre-color balanced, which allows StarTools to make the most of it - great!

Now bear in mind that a lot is also a matter of taste.

Here goes;

--- Auto Develop
Default parameters.
To see what we got. StarTools shows the most pressing issues, being the stacking artifacts and the blue bias. Data looks pretty good. At this stage it's still a bit hard to tell if the data is oversampled or not (e.g. whether binning would be a better option).
--- Crop
First I crop away the artifacts.
Parameter [X1] set to [26 pixels]
Parameter [Y1] set to [113 pixels]
Parameter [X2] set to [4174 pixels (-116)]
Parameter [Y2] set to [2701 pixels (-155)]
--- Wipe
I use Wipe with default parameters to get rid of the bias. As we're still at full resolution, I'm bumping up [Dark Anomaly Filter] to [3 pixels]; this will catch any dead/dark pixels if they're present, so that it doesn't interfere with Wipe's attempt to find the local background.
You can use the Temporary AutoDev feature to see the result. Pretty good I'd say! If you're worried about removing faint nebulosity, you can mask out the bubble if needed (I didn't bother with that).
--- Auto Develop
I launch AutoDev again for the final stretch.
I can now see that at 100% resolution the image is a bit blurry; a combination of atmospheric (seeing) and (possibly) optical limitations. This is very common (in fact it's more often than not the case with DSLR data). If 'smaller'/fainter stars aren't pin points that have their light mostly concentrated in a 3x3 area, then your image is probably oversampled.

The nice thing about oversampled data is that we can use the multiple-samples-per-real-unit-of-resolved-detail to create a better estimate for the real detail (e.g. improve the Signal-to-Noise Ratio without losing real detail), using binning. All this will result in a smaller image (without loss of real detail) that has less noise and, hence, can be pushed harder (for example by stretching more, or by using deconvolution).
--- Bin
I settled for binning to 25% of the original resolution, coalescing roughly every 2x2 pixel block into a new pixel reading that is 4x as accurate.
Parameter [Scale] set to [(scale/noise reduction 50.00%)/(400.00%)/(+2.00 bits)]
--- Auto Develop
I now return to AutoDev again for my final stretch.
I click & drag a Region of Interest over the bubble and surrounding area, as I feel it's the main subject of this image. This will create a global stretch that is optimised for showing exactly that area,
I allow the rest of the image to influence the stretch about 5% (Parameter [Outside ROI Influence] set to [5 %])
--- Deconvolution
I let decon create a mask automatically.
With parameter [Radius] set to [2.3 pixels], we're able to dig out a small bit of extra detail in the bubble. We'll take what we can get! :)
--- Wavelet Sharpen
Using the same mask decon created, I apply some wavelet sharpening to bring out the structures in the bubble a bit more. It's totally to taste.
Parameter [Amount] set to [250 %]
--- Color
This is (IMHO) an important one that gets overlooked a lot; final colour calibration. It is preferably performed towards the end of the work flow.
The colours ST comes up with by default are very close to ideal. Things to look out for is a nice even distribution of all star temperatures across the field (from red->ornage->yellow->white->blue).
Using the MaxRGB button we can see a slight over-representation of green (green should rarely be dominant), so I dial it back a little.
I also introduce more color in the fainter parts, as your data can take it; color noise is fairly low.
Finally, I dial back the saturation to 100%.
Parameter [Dark Saturation] set to [9.80]
Parameter [Saturation Amount] set to [100 %]
Parameter [Green Bias Reduce] set to [1.49]
Parameter [Red Bias Reduce] set to [1.35]
--- Life
The star field is quite busy and can obscure fainter large scale structures in the background.
The Life module's Isolate preset can help in such cases.
--- Wavelet De-Noise
(switching Tracking off)
Final noise reduction to taste.

That's it! And this is what I ended up with;
bubble.jpg
bubble.jpg (1002.52 KiB) Viewed 6121 times
Hope this helps...
Ivo Jager
StarTools creator and astronomy enthusiast
MrMxyzptlk
Posts: 23
Joined: Tue Oct 14, 2014 9:46 am

Re: Bubble Nebula

Post by MrMxyzptlk »

Thank you very much, looks very good, with much better colors than mine :) When I will get home from work I will take a look myself based on your process. I never used the bin function.
One more question: is it possible to remove the star reflection around the brighter stars somehow beautifully (because od the misaligned filter)?
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Bubble Nebula

Post by admin »

MrMxyzptlk wrote:One more question: is it possible to remove the star reflection around the brighter stars somehow beautifully (because od the misaligned filter)?
It's tricky. You might be able to accomplish something using the Filter module... :think:
Ivo Jager
StarTools creator and astronomy enthusiast
Post Reply