Another Rosette - WO Star 71

[gnomus]

Last night was the first clear night in a while.  I managed to get 10 x 20 mins of Ha on the Rosette Nebula with my WO Star 71 + Atik 383.  I would have got more were it not for the usual 'incidents'.  The best of these was around meridian flip time.  SGP asked me if I wanted to bring the flip forward since I was asking it to do a 20 minute exposure and the flip was only 6 minutes away.  The flip worked beautifully and SGP then went about re-centring the object.  Once it had 'solved' the image it then slewed the mount ..... back to the other side of the meridian!  After some head scratching I decided just to carry on with the exposure on the 'pre-flip' side.  The Mesu can (I believe) track 3 hours past the meridian .......   if you have it set up right.  Of course with my technical abilities I am unable to set it up right.  With one minute to go of the 20 minute exposure, the Mesu stopped tracking, PHD lost the star and SGP shut down the sequence!  There may have been some swearing.  

It all worked well after that until the clouds rolled in around midnight. 

I may get some more Ha data but I thought I would process what I have so far.  I used a new PixInsight noise reduction script that I learned about last night when reading a thread by Barry Wilson https://stargazerslounge.com/topic/262356-ngc2244-rosette-in-ha-popular-and-lovely/.  I shamelessly copied Barry's settings (since he also used an 8300 chip, and I haven't a clue how this script works).  There was a definite improvement in noise.  After the script, I did DBE, a 10-iteration deconvolution, and a Local Histogram Stretch (which looked better than the Masked Stretch to my eyes).  Then into PS for some very mild sharpening, a smidge of despeckling, and image size reduction for the purpose of posting here. 

I wanted to have the Rosette slightly off-centre to try to accommodate more of the 'traily' bits of dust I have seen in images by Barry and others.  Another couple of hours of Ha needed perhaps?  (I seem to have got rid of the 'horizontal line' issue I was having previously - https://stargazerslounge.com/topic/260450-horizontal-lines-in-flats/#comment-2851965, so I am pleased about that.) 

 

       

 

[swag72]

That looks very nice Steve - I reckon that you could contrast boost it though as the actual nebula can really benefit from some separation between the bok globules etc.

[gnomus]

2 minutes ago, swag72 said:

That looks very nice Steve - I reckon that you could contrast boost it though as the actual nebula can really benefit from some separation between the bok globules etc.

That's very kind of you to say so.  A bit of curves in PS you reckon, or back to PI?

[Knighty2112]

Very nice image. Last night was great for viewing here too, but alas I was to zonked out from a busy week at work to enjoy it much, except through a quick session with my binos.

[Martin Meredith]

Beautiful image, really sharp throughout. The WO Star 71 looks to be a great scope.

Martin 

[swag72]

13 minutes ago, gnomus said:

 A bit of curves in PS you reckon, or back to PI?

PI? Oh God save me!!! PS of course LOL!!!!

[Mr Spock]

Great looking pic. There's lots of information in there.

[Davey-T]

Nice Rosette Steve , had the Star71 QSI683  out last night but running out of suitable targets now, got mine on side by side triple mounted with WO110FLT and Altair 60mm guide scope so bit of a handful to do a flip and past the Meridian I have S'London LP.
On the bright side it did 20min subs, shame the weather has been so bad that I've missed Orion, Rosette etc for another year, will have take some Jupiters and galaxies now until something big hoves into view

Dave

[Barry-Wilson]

Excellent Steve. I've not seen a poor image from the WO Star 71.

I can imagine the words used regarding the auto meridian flip!  Happy to have a PM re settings in SGP about this.

I agree with Sara, more contrast will help with the 3-d effect. My chosen weapon of choice in PI would be to copy the image, apply an inverted star mask to protect star cores, and then use HDR Wavelets, followed by LHE and an inverted Lum mask to protect the background.  Then blend this enhanced image with the original to your taste (PixelMath or Blend script). Lastly I would then run the Dark Structure Enhance script (experiment with %) and maybe some tweaks with Curves. (Always good to give some PS/PI balance don't you think? LOL!  ).

Using the Mure DeNoise script (referencing your question from my Rosette thread), I didn't apply a mask to protect the high signal areas.  The tutorial didn't mention using one and I simply followed the PI forum article.  I need to refine the measurement of Gaussian Noise using the specific script, but similarly to Sara, after trying and not getting sensible numbers, my head exploded!  Your post Sara is still making me chuckle!

HTH.

[Pompey Monkey]

1 hour ago, gnomus said:

Last night was the first clear night in a while.  I managed to get 10 x 20 mins of Ha on the Rosette Nebula with my WO Star 71 + Atik 383.  I would have got more were it not for the usual 'incidents'.  The best of these was around meridian flip time.  SGP asked me if I wanted to bring the flip forward since I was asking it to do a 20 minute exposure and the flip was only 6 minutes away.  The flip worked beautifully and SGP then went about re-centring the object.  Once it had 'solved' the image it then slewed the mount ..... back to the other side of the meridian!  After some head scratching I decided just to carry on with the exposure on the 'pre-flip' side.  The Mesu can (I believe) track 3 hours past the meridian .......   if you have it set up right.  Of course with my technical abilities I am unable to set it up right.  With one minute to go of the 20 minute exposure, the Mesu stopped tracking, PHD lost the star and SGP shut down the sequence!  There may have been some swearing.  

It all worked well after that until the clouds rolled in around midnight. 

I may get some more Ha data but I thought I would process what I have so far.  I used a new PixInsight noise reduction script that I learned about last night when reading a thread by Barry Wilson https://stargazerslounge.com/topic/262356-ngc2244-rosette-in-ha-popular-and-lovely/.  I shamelessly copied Barry's settings (since he also used an 8300 chip, and I haven't a clue how this script works).  There was a definite improvement in noise.  After the script, I did DBE, a 10-iteration deconvolution, and a Local Histogram Stretch (which looked better than the Masked Stretch to my eyes).  Then into PS for some very mild sharpening, a smidge of despeckling, and image size reduction for the purpose of posting here. 

I wanted to have the Rosette slightly off-centre to try to accommodate more of the 'traily' bits of dust I have seen in images by Barry and others.  Another couple of hours of Ha needed perhaps?  (I seem to have got rid of the 'horizontal line' issue I was having previously - https://stargazerslounge.com/topic/260450-horizontal-lines-in-flats/#comment-2851965, so I am pleased about that.) 

 

      

 

That's a super image Steve (if I may call you that)!  I also agree that it will handle a contrast push nicely which will bring up the more obscured dust globues.

I've got an SBIG STF8300M and I'm off to collect my Star 71 on Saturday - so now I know what to expect in terms of fov - the online calculators never really help me visualise this kind of thing.  

As for running past the meridian, my pier-mounted AZEQ6 will go all the way round for objects below about 40 degrees declination. I've had it running four hours past the big stop sign in the sky before now and the guiding's been as good as on the other side of the pier.

[sloz1664]

Had my WO Star 71 out last night also on the Rosette. But the astro imaging gremlins caught up with me and I had issues with my autofocuser, so I had reset all my offsets.

Good news it's going to be clear again tonight

BTW - Steve Very nice image. Glad to see your WO is performing well after the issues you had with the Tak.

Steve

[gnomus]

1 hour ago, Barry-Wilson said:

Excellent Steve. I've not seen a poor image from the WO Star 71.

I can imagine the words used regarding the auto meridian flip!  Happy to have a PM re settings in SGP about this.

I agree with Sara, more contrast will help with the 3-d effect. My chosen weapon of choice in PI would be to copy the image, apply an inverted star mask to protect star cores, and then use HDR Wavelets, followed by LHE and an inverted Lum mask to protect the background.  Then blend this enhanced image with the original to your taste (PixelMath or Blend script). Lastly I would then run the Dark Structure Enhance script (experiment with %) and maybe some tweaks with Curves. (Always good to give some PS/PI balance don't you think? LOL!  ).

Using the Mure DeNoise script (referencing your question from my Rosette thread), I didn't apply a mask to protect the high signal areas.  The tutorial didn't mention using one and I simply followed the PI forum article.  I need to refine the measurement of Gaussian Noise using the specific script, but similarly to Sara, after trying and not getting sensible numbers, my head exploded!  Your post Sara is still making me chuckle!

HTH.

Thanks Barry

I may take you up on your offer to PM about the meridian flip issue in SGP, if you don't mind.

I used the Mure script without a mask and before DBE.

I followed your advice re: HDR MDT and LHE in PI.  I often find this quite coarse, but the idea of blending with the original was good.  I couldn't find the blending script in PI (I had it before the update).  Anyway I simply exported the TIFF and blended (just over 50%) of the HDR'ed image on top of the original in PS.  Finally, back to PI for Dark Structure Enhancement.   Is it too contrasty now?  

[PhotoGav]

Ooo, that's looking lovely now with the latest raft of tweaks.

I have enjoyed reading through this thread - PI vs PS, which is better? There's only one way to find out...... FIGHT! (In the words of Harry Hill). I was completely lost by Barry's explanation of what to do in PI. I'm with PS, though that's only because I have never dared download PI! And then the marvellous WO Star71 - big fan. I was out last night starting a new project with mine, a wide view of the Jellyfish. Just need a load more really crisp and dark nights to fill up the wells.

 

[swag72]

I like that tweaked edit Steve - Bang on the money

[laudropb]

Very nice detailed image.

[gnomus]

Thanks all.

[Ken82]

Brilliant image thanks for posting

[gnomus]

I managed to get 4.5 hours of RGB last night as 9x10 mins for each channel.  I had hoped to get a bit more Ha tonight but Clear Outside mocked me once again.  So instead I played around with the colour data.  So this is 3 hrs 20 mins plus 1.5 hours each of RGB:

[gnomus]

The previous version ended up way too pink I think.  The perils of processing too late at night.  In light of some other versions I have seen recently, I had another go:

[MartinB]

Despite all your meridian flip problems you've got a great set of data.  I do think you are pushing it too hard though.  Good data doesn't really need all those PI routines.  Just a simple stretch and a little controlled sharpening would probably be a bit easier on the eye.  

[Davey-T]

I'd stop now Steve before you end up in a padded  room

Dave

[Pompey Monkey]

18 minutes ago, MartinB said:

Despite all your meridian flip problems you've got a great set of data.  I do think you are pushing it too hard though.  Good data doesn't really need all those PI routines.  Just a simple stretch and a little controlled sharpening would probably be a bit easier on the eye.  

Agreed - after all the universe is just patterned "noise" and nebulae are extremely diffuse gas clouds.

In my opinion, the eye and brain can do way more for the interpretation of the initial mono image than an arbitrary algorithm in a processing package. I guess that's why all my favourite images still have a hint of noisiness in them  

The mure de noise script in PI is about as far as I'd like to take it. 

[ollypenrice]

I love the Ha. When I use LHE  (or Noel's Local Contrast Enhancement) I then make it a layer over the original and blend in Ps, so, yes, I might have stopped a tad short of the point you reached in the second version. I like what it's done to the main nebula but it's made the faint stuff and the outlying sky a bit harsh to my eye. A useful technique for applying a contrast-enhanced version is to do so in blend mode lighten, where the brightening is applied but not the darkening. (The darkening often brings up background sky noise.) You can also erase parts which are over-cooked.

For me the colour, ideally, might lie between the two you've posted. To bring out more colour variety in the main nebula I think you need either more blue and green signal or an OIII layer. I'm in exactly the same position with mine.

Once again the Star 71 rattles the lion's cage....

Olly

[gnomus]

Thanks again for the response.  This is the sort of criticism I am looking for.  (Davey-T: too late, I've had to go into that padded room several times - quite literally I might add!!!!!)

As a relative beginner, I find it difficult to know when to 'stop' processing.  Every little tweak seems to reveal another puff of dust somewhere, but at some point the image tips over into a bit of a mess.  In light of the response, I wound back on the processing of the RGB image considerably.  I also added less of the Ha into the red channel in 'Lighten' mode (I think it was around 30% from memory), I added around 5% of Ha to the Blue channel, and I dialled back considerably on the Ha as luminance.  I hope this is a little more subtle:

  

[Martin-Devon]

This latest version looks spot-on to me Steve - it's got good contrast without being over-pushed. I always like to see some transparency coming through parts of the nebulosity in the Rosette and you've certainly achieved that. The toned-back palette is also very effective here, nice work. 

Martin