The Monkey Head Nebula NGC 2174 now in colour!

[steppenwolf]

The Monkey Head Nebula

 

The Monkey Head Nebula is an HII emission nebula located 6400 light years away in the constellation of Orion. The nebula is associated with an open cluster of mag. +6.8 located at its centre. With a diameter of approximately 40 arcminutes, the nebula is larger than the full Moon. There is much debate over the correct designation of this object as some sources cite the nebula as being NGC 2174 and some stating the nebula as being NGC2174/5. What is not in contention is that Stewart Sharpless logged this object as SH2-252 in his second and final catalogue completed in 1959. Personally, I go for the nebula being NGC 2174 and the open cluster being NGC 2175!

It was and I guess still is, my intention to produce a bi-colour version of this image but the skies have not been kind and just the Ha displayed here has taken 5 difficult nights of cloud-dodging although in fairness, it has been CCD Commander and my automation project that have done most of the starting and ending of the sessions! I decided that with the weather currently deteriorating, I may as well post up the project thus far.

Image Stats
Mount: Mesu 200
Telescope: Sky-Watcher Esprit 150
Flattener: Sky-Watcher Esprit specific
Camera: QSI 683 WSG-8
Filter: Astrodon 3nm Ha
Subframes: 22 x 1800 sec Ha
Integration: 11 hours
Control: CCD Commander
Capture: MaxIm DL
Calibration, Stacking and Deconvolution: PixInsight
Post-Processing: PhotoShop PS3

Full Resolution version

Location of the open star cluster NGC 2175

Visually

This target lies 2.3° to the southeast of mag. +3.3 Propus and is a two for one object, comprising an open cluster (NGC 2175) and an emission nebula (NGC 2174) although most astronomers associate the nebula with NGC 2175. Lying in a rich star field, the almost circular shape is punctured by a dent in its western edge which stands out particularly well in images and helps to form the ‘monkey-head’ shape from which the nebula gets its common name. Inboard of this dent towards the east is an area of intense star birth. You’ll need a very large telescope to discern the shape of the nebula but a 4- 6-inch telescope will show the open cluster very well.

Location of Nebula and associated Star Cluster

RA: 06h09m 36.0s DE:+20°29'00"

Star Birth Region

On the eastern limb there is a dent pushing westward into the nebula and here there is a huge stellar nursery that featured in the Hubble Space Telescope’s IR image (http://hubblesite.org/news_release/news/2014-18) released on 17th March, 2014. The nebula is comprised mainly of hydrogen gas which is ionised by the intense ultraviolet radiation emitted by the hot young stars within the nebula. This ionisation causes the hydrogen to glow red. Also associated with this nebula are some faint regions of reflection nebulosity giving a hazy blue appearance and some relatively faint dust lanes add interest to the interior.

 

 

 

[swag72]

You've beaten me to it  

Looks OK in Ha! Also you can get it all in one frame.... I've had to mosaic it  I love that close up star birth region .....  

[iansmith]

Very nice image.

Cheers, Ian

[Scott]

damn fine image mate. so much detail.

[steppenwolf]

13 hours ago, swag72 said:

Looks OK in Ha! Also you can get it all in one frame.... I've had to mosaic it  I love that close up star birth region .....  

Thanks, Sara, the Esprit 150 was born for an image at this size and Ha suits it very well indeed although if there is time (oh yeah, really?) I want to do an LRGB version as well!

That star birth region is amazing and I used it as my guide during deconvolution.

12 hours ago, iansmith said:

Very nice image.

Thanks, Ian

12 hours ago, Scott said:

damn fine image mate. so much detail.

Thanks, Scott as you know from the preview you saw of this image, the extra processing work I put in was well worth it to release that detail.

[MikeODay]

Lovely image and very interesting write-up.  Thanks for sharing.

[steppenwolf]

1 hour ago, MikeODay said:

Lovely image and very interesting write-up.  Thanks for sharing.

Thanks, Mike, pleased you like it!

[laudropb]

Great image and a very informative write up. Must revisit this and try for the open cluster. Thanks

[Petergoodhew]

A beautifully crisp clear monkey head Steve.  Makes me want to have a go if the clouds will leave me alone one day. That comprehensive write-up puts my minimalistic posts to shame. Peter

[cfpendock]

Good stuff - as usual from this source.  Especially with the description of what exactly is in the image.  I like it.

Chris

[steppenwolf]

19 hours ago, laudropb said:

Great image and a very informative write up. Must revisit this and try for the open cluster. Thanks

Thanks, it certainly is worth hunting down the cluster itself.

5 hours ago, Petergoodhew said:

A beautifully crisp clear monkey head Steve.  Makes me want to have a go if the clouds will leave me alone one day. That comprehensive write-up puts my minimalistic posts to shame. Peter

The write-up is a smoke screen for the lack of colour content although I am just about to rectify that with the 11 x OIII subframes I snatched last night! I have to do a lot of cloud-dodging and even last night's session lost me an hour's work due to high,thin cloud!!!

5 hours ago, cfpendock said:

Good stuff - as usual from this source.  Especially with the description of what exactly is in the image.  I like it.

Thanks, Chris, pleased you like it.

[steppenwolf]

To my astonishment, I got another clear night last night and managed to capture some OIII data  (11 x 1800 seconds) for this object so here is the next stage, an Ha/OIII/OIII image.

[kirkster501]

That’s very nice indeed Steve. Wow, long exposures at 1800s in the ??!

[MarsG76]

Very nice image... Nicely balanced levels.

[Barry-Wilson]

Very good Steve, wonderfully sharp (but not over sharpened) details in the nebula.

[steppenwolf]

3 minutes ago, Barry-Wilson said:

Very good Steve, wonderfully sharp (but not over sharpened) details in the nebula.

That's interesting that you should say that, Barry as I haven't applied any 'sharpening' (in the normal sense) to this data at all but I have deconvolved it in PI.

[Barry-Wilson]

11 minutes ago, steppenwolf said:

That's interesting that you should say that, Barry as I haven't applied any 'sharpening' (in the normal sense) to this data at all but I have deconvolved it in PI.

I guessed that to be the case Steve, hence the comment in brackets.

Deconvolution is a wonderful tool in the processing kit bag when you are presented with substantial structure to 'sharpen'; used here in the colloquail sense, not optical image processing sense.  It's easy to over deconvolve and your image is a good example of the technique - tightening stars and reducing the effects of imaging and data collection/processing.

That's all I meant rather than implying you had sharpened with UnSharp Mask etc.

There is whole thread emerging here about deconvolution its purpose and techniques . . . sorry not meaning to derail.

[steppenwolf]

3 minutes ago, Barry-Wilson said:

That's all I meant rather than implying you had sharpened with UnSharp Mask etc.

Be assured that I didn't read anything into your comment at all, Barry - what you don't know and couldn't have known is that this is literally the first image that I have ever deconvolved that I have been happy with but I am finally on top of the process now whereas previously I always found it a bit 'hit and miss' with a whole range of different pieces of software!

[Barry-Wilson]

12 minutes ago, steppenwolf said:

Be assured that I didn't read anything into your comment at all, Barry - what you don't know and couldn't have known is that this is literally the first image that I have ever deconvolved that I have been happy with but I am finally on top of the process now whereas previously I always found it a bit 'hit and miss' with a whole range of different pieces of software!

Deconvolution does take quite a bit of practise I admit.  It helps if you minimise swapping OTAs and ccds so you become accustomed to the stacked results of your optical train . . . this then remains a constant among many variables, seeing and other capture effects etc.  I rarely ever use more than 25 or 30 iterations too, so easy to over do in one's enthusiasm for revealing lost detail with consequential artefacts only becoming evident later in the non-linear stages.  Sometimes, no more than 5-10 iterations are required to just spruce up and 'tighten' the image.

I'm pleased you are satisfied with the results on your image - I felt it certainly showed, hence I thought I had to comment.  Deconvolution is sometimes presented as a mysterious and complex process - now whilst it isn't the simplest of techniques it can be mastered with patience and perseverance.

Good job on your Monkey Head Steve!

[steppenwolf]

3 minutes ago, Barry-Wilson said:

Deconvolution is sometimes presented as a mysterious and complex process - now whilst it isn't the simplest of techniques it can be mastered with patience and perseverance.

I have certainly found it a little 'mysterious' in the past but have to say that the PI approach seems quite sound once you get the hang of 'de-ringing' which is a little trial and error. The key to my recent success with the process was to use DynamicPSF to generate an 'external' PSF image as I found that Parametric PSF gave me the sort of inconsistent results that I have previously had with other deconvolution software.

There are some great 'sharpening' techniques that one can use in PhotoShop and I use them myself but there is something rather satisfying about having a purer linear image available before slipping into PS for the rest of the image processing which is what I prefer to do - PI for linear stages, PS for assembly and non-linear processing.

[Petergoodhew]

52 minutes ago, Barry-Wilson said:

There is whole thread emerging here about deconvolution its purpose and techniques . . . sorry not meaning to derail.

I think a thread on this, and sharpening, would be very valuable to many of us.  It's hard to know how much is enough, and when it's too much. I'm sure that I overdo it often, but also have had some images lacking in detail that was lurking in among the photons.

[Filroden]

27 minutes ago, steppenwolf said:

once you get the hang of 'de-ringing' which is a little trial and error

If you ever solve this aspect of de-convolution then please share! I have to randomly lower the dark and I never touch the light settings and I'm never certain why I'm doing it or what will happen until I hit apply to the preview.

But that's a lovely Monkey Head. I love the tricolour palette. Very Bastille Day!

[Barry-Wilson]

34 minutes ago, steppenwolf said:

I have certainly found it a little 'mysterious' in the past but have to say that the PI approach seems quite sound once you get the hang of 'de-ringing' which is a little trial and error. The key to my recent success with the process was to use DynamicPSF to generate an 'external' PSF image as I found that Parametric PSF gave me the sort of inconsistent results that I have previously had with other deconvolution software.

Exactly - this is one of the key steps.  Not to be over-zealous or narcisstic about this, I do have a Deconvolution tutorial on my SmuMug site, here.  I have also refined the PSF selection technique with a further step, it improves the end-result if used but not excessively so, so the tutorial stands as an effective Deconvolution workflow.  I will get round to updating the tutorial.

It is critical to select stars that are not saturated, relatively small and use only the 'Moffat' function sizes, ie not Gaussian or Lorentzian.  Choose about 40 - 50.  Definitely black-clip the Mask and carefully use the StarMask.  It is also important to use a large preview window - too small and it will not accurately reflect the results on the whole image.

HTH.

I'm more thann happy for people to create a new thread so we do not disrupt Steve's on his excellent MonkeyHead . . . and derail more than we have done so already.

[SilverAstro]

Great image and brill. write-up, thanks, most interesting.

[steppenwolf]

49 minutes ago, Petergoodhew said:

I think a thread on this, and sharpening, would be very valuable to many of us.  It's hard to know how much is enough, and when it's too much. I'm sure that I overdo it often, but also have had some images lacking in detail that was lurking in among the photons.

I think a new thread would be cool and if anyone is interested I'll happily produce a copy of my workflow as used for this image although I think that I am right on the cusp of going too far with this one but it satisfied my desire to get that star-birth area nice and tight!

44 minutes ago, Filroden said:

If you ever solve this aspect of de-convolution then please share!

I think it really is a trial and error exercise as each integrated image will have a potentially different star profile - I had to make changes between my Ha and OIII images for this object for example although in my case it was just a case of reducing the Global Dark setting in my case.

44 minutes ago, Barry-Wilson said:

It is critical to select stars that are not saturated, relatively small and use only the 'Moffat' function sizes, ie not Gaussian or Lorentzian. 

Absolutely, I chose 30 stars but removed the 3 that were Gaussian but you pretty get used to what stars will stand up to selection and those that are not going to be suitable. Thanks for the link, I shall have a look as I suspect I won't need to re-invent the wheel with my own workflow!

46 minutes ago, Barry-Wilson said:

Definitely black-clip the Mask

I didn't get on very well with this step as it seemed to attenuate the deconvolution too much but although I am confident with PS masks, I am not up to speed with PI ones!