Tips for improving DSLR LMC?

[rogerg]

Hi all,

On the weekend I captured this LMC shot. The data is quite nice, taken at f/2.8, 120s exposures in good conditions.

As you can see it has that blue low-Ha DSLR look.

Any tips on improving this?

I've tried selective saturation and levels, and just the usual results are achieved. No luck pulling out any colour variation.

I've processed all the weekend's photo's in PixInsight with reasonable success using ImageCalibration, ImageRegistration, ImageIntegration and DBE. Limited success with Histogram, and have taken the image in to Photoshop for any further processing as I'm so comfortable with Photoshop's tools as compared to PI. I'm open to suggestions in PI thought, and have tried tips in some of Harry's video on this, but no luck so far.

Dropbox link to 1500px image 1.1mb unable to be uploaded here:

https://www.dropbox.com/s/lyhd9526fh8u9pz/Large-Magellanic-Cloud-%28100mm-f2.jpg

Regards,
Roger.

[Octane]

Hi Roger,

Mate, I think that's about as good as it gets with a DSLR (unmodified).

I've got a shot taken from 6 years ago with a 300mm lens and it's the same monotone through the middle.

You should be chuffed with this; it's really quite visually-appealing.

H

[rogerg]

Wow, thanks H, it didn't even occur to me I'd get a complement on the photo :lol: I'm pleased to hear it's comparable to your results.

[strongmanmike]

I agree with H even with the muted Ha this is quite a good shot Roger

You and colour...are pretty well the opposite to me and colour :P :lol: all good though, like your work :thumbsup:

Mike

[rogerg]

Haha :lol: yes you have a good point there Mike, even with a OSC DSLR, my astro shots come out muted colour :lol: Thanks :)

[gregbradley]

I agree that is all the data you collected. Ha was filtered out by the low pass filter of the camera. I looked at my DSLR shots and they look the same.

Greg.

[rogerg]

Thanks Greg.

[RickS]

You can dig a bit more colour out by steepening the a* and b* curves in LAB mode (either in PS or PI) but it's remarkable how little colour is in that image, Roger! I have gone a bit over the top here... ;)

Cheers,
Rick.

[rogerg]

:) Thanks for trying Rick!

[rcheshire]

Mind if I have a go?

[alpal]

Hi Roger,
I had a go too.
There is something awfully wrong with the colours.
When I boosted them in LAB mode all the Tarantula nebula etc went blue!
How could that have happened?

I ended up separating channels & swapping blue with red
- then using LAB mode to get the nebulas coming out red.

Does this look better?

[Peter.M]

The blue comes from the o3 and h-beta that is collected by the dslr. The ha is stronger than the hb so in essence what you are doing should be ok for the nebula but less ok for the stars.

[alpal]

Ok thanks -
so the O3 & H-beta are blue & the DSLR has picked them up
in preference to the red of Ha.

[MrB]

Have always had the same sad issue with the 60Da.
The Da's Ha response is only marginally better than that of a non-modified DSLR, but I would have thought it'd show a little hint of red, but nope.
I had suspected the Hb may be the culprit but thanks to Peter I now know for sure!

I might have to try a Ha + OSC blend... tho I've never done this before, is it possible? Would you blend the Ha into the OSC red channel?

[alpal]

Why not cheat like I did & swap the red & blue channels over?

[rogerg]

Thanks all for contributing, it's been great to see other people's attempts and hear comments on this:thumbsup:. At least I don't feel so bad about my attempt now :)

I hope no one minds, but in the end, of the attempts made, I think I like my original the best. I like the variety of colour in alpal's and Rick's, but in the end it's not the pink nebula's which are turning up pink. Rowland's is showing great fainter extends of the LMC coming out, but not the colour variation. Fact is, data isn't there to split the colours it seems.

I thought I had seen in the past unmodified DSLR attempts on such objects which surprised me as to their colour variation and pink nebulas. I think perhaps in the end they were using either modified DSLR or some combination of DSLR and specific spectrum data. It'll be interesting if I come across them again.

[ourkind]

It looks great Roger :thumbsup:

[rcheshire]

CLS-CCD filter from Astronomik. Exposure times with an unmodded camera extend to 5 or more minutes. Great if you have guiding.

[rogerg]

Well, I have applied some of Justin's magic (http://www.iceinspace.com.au/forum/showthread.php?p=1070687) :)

I dunno, perhaps better? probably better. Even if it is using a little "trickery" :)

[jsmoraes]

Perhaps ?

Bright and contrast in channels color. Layer mask on parts of galaxy to enhance red color in curves. Other layer mask to get some contrast of dark areas with contrast. Filter Minimum in shadow mask layer with 20 % opacity to reduce the presence of background stars.

note: My experience says that, with DSLR Canon 350D, some photos with High ISO cause saturation of colors to white in stack operation. Some times I prefer more number of photos with low ISO - (800 or 400)

[alocky]

Hi roger -same experience here with the Nikon d800
As others have said you really need a modded dslr. Funny thing is that there's plenty of other dso's with dominantly h-alpha emission that turned out fine with the D800, but I've never managed to get a red tarantula without using a mono CCD and lrgb filters.
The LAB processed version you've produced is pretty damn good.
With a 180mm
http://www.astrobin.com/full/62545/B/
Through the tak FSQ and the best I could get was a sort of dried blood tinge.
http://www.astrobin.com/27741/
Cheers,
Andrew.

[jsmoraes]

Undoubtly with filters LRGB and H alpha we get more real image of DSOs. No discussion about it. But some color information can be captured with DSLR. The seeing, black sky, and graphic procedures are factors with more influence in the result.

ISO 400 - canon 350d - 18 x 120 sec - http://astronomia-e-astrofotos.1069742.n5.nabble.com/file/n1288/ngc2070-131210-1k.jpg

ISO 800 - canon 350d - 20 x 120 sec - Skyglow filter - http://astronomia-e-astrofotos.1069742.n5.nabble.com/file/n1359/ngc2070-140130-k7.jpg