IC1848 Soul Nebula (HST)

[strutsinaction]

Hi

It's been a while since I've imaged anything but recently I was able to grab a few IC1848 subs, despite the presence of the moon (>50%). Subs were captured on consecutive nights with the moon getting gradually brighter. The Ha subs were very clean but both the OIII and SII subs were quite noisy, probably due to the presence of the moon.

Details:

Ha - 15x600s
OIII - 15x600s
SII - 15x600s

QSI690
Astrodon filters
Avalon Linear
Tak FSQ85 + reducer (f3.9)
SGPro
PixInsight
PhotoShop (Selective Colour to get the HST mapping)

I'll have to redo the OIII an SII subs at some point without the moon to see if that improves the final image.

Thanks for looking. Any comments/critique welcomed!

Regards
John

[D4N]

That's looking really nice.

I hear what you're saying with the OIII and SII, it's frustrating to take one hours Ha and it be far better quality data than two hours OIII and two hours SII.

[Rodd]

Looks good.  In Pixinsight--use the Mure denoise script--works amazingly well if you don't use Drizzle integration.  Also--remove the first (and maybe second) wavelet layers with MSLT, MSMT or the older Atrous Wavelets.  That will pretty much remove the noise--then you can regain the detail by selective sharpening using masks (Range mask, or I like to use pixel math and subtract a star mask from a range mask--that keeps the stars and background covered).

Great image.

[strutsinaction]

23 minutes ago, Rodd said:

Looks good.  In Pixinsight--use the Mure denoise script--works amazingly well if you don't use Drizzle integration.  Also--remove the first (and maybe second) wavelet layers with MSLT, MSMT or the older Atrous Wavelets.  That will pretty much remove the noise--then you can regain the detail by selective sharpening using masks (Range mask, or I like to use pixel math and subtract a star mask from a range mask--that keeps the stars and background covered).

Great image.

Hi Rodd

Thanks for your comments. MureDenoise is my weapon of choice for noise reduction. It's fantastic and certainly helped with my noisy OIII and SII subs but I pushed it as far as I could without introducing additional artifacts. I didn't consider applying any further noise reduction so that's something I could try.

In summary, the PixInsight processing steps I applied were:

- MureDenoise to Ha, OIII and SII
- DBE to Ha, OIII and SII
- LinearFit (OIII and SII -> Ha)
- ChannelCombination
- BackgroundsNeutralization
- HistogramTransformation
- MorphologicalTransformation (star reduction)
- MagentaStarReduction
- MultiscaleMeadianTransformation (sharpening)
- CurvesTransformation

Regards
John

[Rodd]

2 minutes ago, strutsinaction said:

Hi Rodd

Thanks for your comments. MureDenoise is my weapon of choice for noise reduction. It's fantastic and certainly helped with my noisy OIII and SII subs but I pushed it as far as I could without introducing additional artifacts. I didn't consider applying any further noise reduction so that's something I could try.

In summary, the PixInsight processing steps I applied were:

- MureDenoise to Ha, OIII and SII
- DBE to Ha, OIII and SII
- LinearFit (OIII and SII -> Ha)
- ChannelCombination
- BackgroundsNeutralization
- HistogramTransformation
- MorphologicalTransformation (star reduction)
- MagentaStarReduction
- MultiscaleMeadianTransformation (sharpening)
- CurvesTransformation

Regards
John

Did you forget to do Color Calibration--or did you just forget to write it down?  Colors look good so I'd say the latter.  I am starting to really like PI--it took a while though, and allot of hard work.  It's usually the little things--like does the mask really cover want you want it to.  You know the drill--if you can produce an image like this!

[swag72]

That's looking very nice indeed - I like it

[ollypenrice]

Good to see this take centre stage. It really is a lovely nebula in its own right. A most attractive rendition.

Olly

[Gina]

Lovely

[Jokehoba]

Terrific image. Just grabbed some Ha on this myself a couple of nights ago (despite a nearly full Moon). I was going to leave it at that, but your image is spurring me on to collect some OIII data. Unfortunately I don't yet have an SII filter...but maybe it's time to get all the colours. 

[Twisted Lip]

I reckon that is stunning, has a 3D quality to it. Top top image!

Will

[strutsinaction]

3 hours ago, Rodd said:

Did you forget to do Color Calibration--or did you just forget to write it down?  Colors look good so I'd say the latter.  I am starting to really like PI--it took a while though, and allot of hard work.  It's usually the little things--like does the mask really cover want you want it to.  You know the drill--if you can produce an image like this!

Hi Rodd

I don't apply ColorCalibration to purely narrowband images. I've been led to believe it's not necessary as the colours are false. I wonder what others think?

Regards
John

[Rodd]

1 minute ago, strutsinaction said:

Hi Rodd

I don't apply ColorCalibration to purely narrowband images. I've been led to believe it's not necessary as the colours are false. I wonder what others think?

Regards
John

Try it--I think you will like it.  I do--always improves my color balance--less playing with curves after.  PI was designed for NB as well you know.

[ollypenrice]

5 minutes ago, strutsinaction said:

Hi Rodd

I don't apply ColorCalibration to purely narrowband images. I've been led to believe it's not necessary as the colours are false. I wonder what others think?

Regards
John

Maybe a matter of getting a neutral background sky? I'm not a colour-mapping imager so I don't know.

Then again, once above the background I'd have thought that the idea was to map the colours as closely as possible to their measured intensity?

Hey, what do I know? Nothing!

Olly

[strutsinaction]

16 minutes ago, Rodd said:

Try it--I think you will like it.  I do--always improves my color balance--less playing with curves after.  PI was designed for NB as well you know.

Hi Rodd

I'll have a play and see if it makes any difference. PixInsight makes it easy to experiment!

Regards
John

 

[strutsinaction]

18 minutes ago, ollypenrice said:

Maybe a matter of getting a neutral background sky? I'm not a colour-mapping imager so I don't know.

Then again, once above the background I'd have thought that the idea was to map the colours as closely as possible to their measured intensity?

Hey, what do I know? Nothing!

Olly

Modesty will get you everywhere 

[Barry-Wilson]

I nice image John.

Unless you deliberately want the teal colour, have you tried some SCNR green?  I often find I need to use it on HST images, really helps with the blue OIII channel.  I sometimes will protect the high signal areas with a luminance mask and using Curves reduce the stauration of the background by dragging the right hand end of the curve down, limiting the whole range.  Using a small gentle ACDNR as NR can recover some room by the black point on the histogram too.

I've downloaded the .png file and adjusted as above to show you - will PM this to you and you can decide whether might like to adjust or not - obviously do as you see fit.  Only making suggestions to help.

Barry

[PatrickGilliland]

16 minutes ago, Barry-Wilson said:

Unless you deliberately want the teal colour, have you tried some SCNR green? 

Nice work - one of my nemesis object plenty of data but never created a version i am happy with.  This is a nice version so well done.

An interesting point Oiii should as i understand it should be teal (even though as per Barry's suggestion i often make blue as it is more widely liked (at least by me)).

As you can see though OIII is actually about halfway between blue and green and 'strictly speaking' that is what colour it should be (roughly teal) it will most likely give the most detail/impact when mapped at that spectral range.  Having said all that i always assign to blue but looking at the math now i may experiment with weighting it to green a little as well.  I might be missing something but may need to be more creative in PI.  If assign OIII to blue channel it targets blue output (has a range set for blue of x nm?) what if i want third channel to be teal not blue.  I can't mix with green as this will have Sii and a little Ha assigned so weighting would be wrong.  Anyway back to the point.....

It should be teal any change is artistic isn't it?

Paddy

[strutsinaction]

38 minutes ago, Barry-Wilson said:

I nice image John.

Unless you deliberately want the teal colour, have you tried some SCNR green?  I often find I need to use it on HST images, really helps with the blue OIII channel.  I sometimes will protect the high signal areas with a luminance mask and using Curves reduce the stauration of the background by dragging the right hand end of the curve down, limiting the whole range.  Using a small gentle ACDNR as NR can recover some room by the black point on the histogram too.

I've downloaded the .png file and adjusted as above to show you - will PM this to you and you can decide whether might like to adjust or not - obviously do as you see fit.  Only making suggestions to help.

Barry

Hi Barry

I really like what you achieved with the background, certainly a lot 'cleaner' than my rendition, but I do prefer the classic gold/turquoise mapping even though it's a little less subtle than your interpretation  

Thanks for your suggestions, most appreciated!

Regards
John

[Stephen Green]

a very nice image and detail