Camera Labs

Hi folks,

I've been housebound this morning (waiting for half a lamb to arrive - yes, think of your own jokes!) so I thought I'd have another go at processing the North American nebula. I used the same full sized image file used to produce the image in my previous post but today I used multiple passes of the PhotoShop "Dust and Scratches" filter to remove all the stars from that image. I found the original tutorial for this here although I see that things have moved on to version 2 now. Anyway, with a version of the image devoid of stars I first saved a copy and then subtracted that image from the original using PhotoShop. The result was an image of just stars. The point of this was that I could then open the "stars" image in PixInsight and use its powerful "Deconvolution" routine to reduce the size of the stars considerably without the usual black halos that can appear in areas of nebulosity - PixInsight rocks. I then saved the deconvolved image and opened it in PhotoShop.

Before I added the stars back I took the opportunity to do some really heavy processing of the nebulosity. That involved multiple passes of noise reduction as well as colour manipulation to emphasise the "red bits". Noel Carboni's "Actions" were also useful at various points. Having got the nebulosity where I wanted it I then did some further manipulation of the "stars only" image to increase the brightness and remove the dimmer stars. After some manual clean up in both images I combined them back together.

I find the difference between the two versions quite startling, though this image certainly owes a certain amount to (an attempt at) artistry than scientific accuracy but I'm pleased with the result as the Hα regions show nicely while the "white bits" have the intended dreamy quality. I certainly prefer this version to the one above so I hope you feel the same way:

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

That's simply stunning, Bob.

_________________
Radu
Canon PowerShot S100
Canon 50D , SIGMA 10-20 f3.5 ,Canon EF 24-105 L IS USM, Canon EF 100/2.8 macro Canon EF 50/1.4 ,Canon EF 85 f1.8,Canon EF-S17-85 4-5.6 Old Tamron 28-300 inherited from my Canon Rebel G film camera
Canon580EXII
http://www.errre.net

Wow great image Bob!

_________________
Camera: Canon 550D with battery grip
Lenses: Canon 24-105mm f/4L, Canon 50mm f1.4, Canon 18-55mm, Tamron 70-300mm,
Accessories: Manfrotto 055XPROB with 808RC4 head, Canon 430ex II speedlite, Lowepro Nova 180AW and Lowepro Pro Runner 450AW

Oh that is so lame, every hot girl who can aim a camera thinks she’s a photographer -Stewie Griffin

Wow, Bob, that's a really interesting test. The second image looks a little unreal to me, but there's no doubt the star images have been kept in check as you've boosted the nebulosity. A really nice application of the technology...

PS - to hijack this thread slightly, popo, do let me know if you ever mod your 450D as I have one here which I could either sell or mod...

Hi folks,

Thanks for the kind words.

@ Gordon What do you mean by trying to hijack my thread!

Seriously, I know where you are coming from. I think the star removal process has resulted in the nebulosity and the surrounding background of very faint stars as one looks down the plane of the galaxy becoming almost indistinguishable. But on the positive side it has allowed the dust lanes to show through much more clearly with that overburden of brighter stars removed.

I purposely didn't refer to any images of the North American nebula while I was processing. As a result I was quite surprised to see that, if Google Images is any guide, there are relatively few ultra widefield shots of the area and, at first glance, even fewer which weren't produced with narrowband filters. Hopefully next summer I'll return to the subject but this time with my ML16803 camera and a combination of broadband and narrowband images to work with. My 165mm Pentax MF lens should provide a similar field of view but I've yet to have any chance of testing that combination.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi folks,

Had another pass at processing the North American Nebula shot above. Different choices can have quite an effect. Here's the latest version:

And, for easy reference, here was the previous effort:

The latest version, at the top, had markedly less noise reduction applied to the nebulosity resulting in a crunchier rendition and I also enhanced the red Hα regions rather more. I also made some different choices on how the stars were added back to the image. I made no attempt to match the white balance of the two versions.

• Which do you prefer?
  
• Any thoughts on how to improve on either?

I have a horrible suspicion that the sweet spot is somewhere between the two. The top one may have a more natural look but I still prefer the bottom one for its ability to show subtle variations in the dust lanes.

This post-processing lark ain't easy.

Update: Hmm, I think I prefer the latest version (top of this post) today (Tuesday)!

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi folks,

Over three years ago I posted a mosaic of the main stars of the constellation of Orion (here). I thought it would be interesting to see how PixInsight coped with what, frankly, was some pretty rough raw imagery and, to my horror, only one dark and one flat frame!!! Please refer to the earlier posts for the shooting information.

There were some severe light pollution gradients to cope with which I had difficulty removing entirely using the "DynamicBackgroundExtraction" tool. Maybe if I spent more time I'd do better but, happily, after "registering" the two halves (PixInsight was far better than the PhotoShop routine I used three years ago) I was able to build the mosaic using PixInsight's downloadable "GradientsMergeMosaic" tool which did a neat job of removing the residual gradients.

I then used PhotoShop's "Dust and Scatches" filter as before to end up with a "No Stars" and a "Just Stars" version of the image. That allowed me to tune the nebulosity processing without having to worry about horrible effects on the stars and also to reduce the star sizes (deconvolution) without worrying about halos. Some manual clean up was still needed on the "Stars Only" image and then I combined the two together so that the stars didn't end up dominating the image.

The result is certainly different and, IMHO, in many ways better but I think my neophyte usage if the background extraction tool may have cost me some of the faint Hα emissions. But on the positive side I was able to considerably reduce the star clutter, extract more detail from the Orion nebula (in the sword) and show more clearly the colours of the Horsehead and Flame nebulae at the left end of the belt.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Bob, you're not helping me with the wait for my converted camera to arrive! The whole Orion area is so rich and full I can just keep looking at it. I can take a little comfort there hasn't been any clear skies recently, and it's a bit early in the season yet but there's other bits to play with until then. This year I really want to give the horsehead and flame a closer look.

Haven't heard of PixInsight before so will have to get a trial of that and run historic data through it and see what I get...

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

To those unsure what the North America nebula region looks like, it's something like this:



I think you need to relook at the processing of this one from scratch Bob.

Greg

_________________
Nexstar 11 GPS, 2 x Sky 90, M25C, MaximDL, Photshop CS3, Noel Carboni's Photoshop actions, 7 foot Pulsar fibreglass dome, Canon 40D, 100mm macro lens, 28-200mm zoom lens, 17-55mm f#2.8 zoom lens, 100-400mm zoom lens, 1.4x converter, 2x converter.
http://www.newforestobservatory.com/
http://www.flickr.com/photos/12801949@N02/

Hi Greg,

Cracking image, as always, and it certainly got me asking myself why there is such a difference! Part of it may be down to your bigger light bucket (280mm diameter mirror Celestron telescope running at f/2 with HyperStar I believe) so that explains the extra detail over a standard DSLR and a 70mm diameter lens at f/1.2. But that doesn't explain the significant colour difference between the two images.

Unfortunately the Sony ICX453AQ in your SXVF-M25C doesn't have a datasheet that I can find but I do note that Starlight Express suggest a quantum efficiency of about 30% at 650nm (approximately the wavelength of Hα) as they quote a 50% roll-off from a peak QE of 60% in the green region. Christian Buil suggests a QE for the 5D2 of 34% in the green dropping to 5% at 650nm - that's just 14% of the green figure.

So I'd like to respectfully suggest that you are being a little unfair. My image had to work with the white balance of the areas distant from the nebula which had no Hα contribution. Where Hα was present my 5D2 was capturing much less of it, as a proportion of the total flux, than your own dedicated astronomical CCD camera and if I try to boost the red much more using PhotoShop's Hue/Saturation dialog I get some very unnatural results elsewhere in the image. A quick Google search for unmodified DSLR images of the region shows, after careful inspection to check that the displayed results actually are from unmodified DSLRs, that my images above, while still in need of improvement, are better than some and worse than others. The best looking results from that search tend, however, to have a narrower field of view concentrating on just the North American nebula itself with sufficient vignetting to allow, IMHO, a degree of cheating regarding the white balance!

But thanks for the feedback - it keeps me honest even if I do have to spend a lot of time researching my responses - and I will try to improve!

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Seeing Greg's output, awesome as always, does make me wonder why I bother since I'm not going to get anything near that without a lot more cost and effort. But still, there's always the thrill of doing the best you can.

For comparison of the North America nebula, I've dug out my attempt from last year:



I hope to do better this year, and today a new toy has arrived which might contribute towards that...

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

It was taken with the Sky 90 Bob - so 90mm compared to your 70mm - not as big a difference as the Hyperstar light bucket
Greg

_________________
Nexstar 11 GPS, 2 x Sky 90, M25C, MaximDL, Photshop CS3, Noel Carboni's Photoshop actions, 7 foot Pulsar fibreglass dome, Canon 40D, 100mm macro lens, 28-200mm zoom lens, 17-55mm f#2.8 zoom lens, 100-400mm zoom lens, 1.4x converter, 2x converter.
http://www.newforestobservatory.com/
http://www.flickr.com/photos/12801949@N02/

.
Fair enough, though the focal length of your Sky 90 certainly helped so far as the fine detail is concerned. But I still think the unmodified DSLR's insensitivity to Hα severely limited what I was able to accomplish, particularly with such a wide field of view and the aforementioned white balance constraints that imposed on the shot. The Milky Way has to look milky, not like a Bloody Mary!

Hopefully in a few months I'll be doing wide field imaging with my just delivered, but as yet untested, ML16803 camera behind my Pentax 165mm f/2.8 medium format lens as well as a soon to be delivered TEC 140 'scope. At that point I'll have totally run out of excuses for poor results!!! While I hope to concentrate on narrowband imaging with both set-ups I'll also be taking RGB shots as I plan to get a set of Astrodon Tru-Balance filters as well as the narrowband ones (Hα and Hβ as well as OIII, NII and, eventually, SII). So next summer I'll try and return to NGC7000 if the bankruptcy court hasn't knobbled me first! Wish me luck.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Best of luck to you Bob
Greg

_________________
Nexstar 11 GPS, 2 x Sky 90, M25C, MaximDL, Photshop CS3, Noel Carboni's Photoshop actions, 7 foot Pulsar fibreglass dome, Canon 40D, 100mm macro lens, 28-200mm zoom lens, 17-55mm f#2.8 zoom lens, 100-400mm zoom lens, 1.4x converter, 2x converter.
http://www.newforestobservatory.com/
http://www.flickr.com/photos/12801949@N02/

Actually - just saw your filter list - why NII?
Greg

_________________
Nexstar 11 GPS, 2 x Sky 90, M25C, MaximDL, Photshop CS3, Noel Carboni's Photoshop actions, 7 foot Pulsar fibreglass dome, Canon 40D, 100mm macro lens, 28-200mm zoom lens, 17-55mm f#2.8 zoom lens, 100-400mm zoom lens, 1.4x converter, 2x converter.
http://www.newforestobservatory.com/
http://www.flickr.com/photos/12801949@N02/