NGC 925 - nice galaxy in Triangulum

[lukebl]

Hi all,

In a rare (for me) session with near perfect guiding, I managed to capture this lovely barred galaxy, which I stumbled across whilst browsing on Carte du Ciel. It doesn't crop up too often, which is odd as it's quite bright and big at 10.5' x 5.9', and with an interesting form. Still getting to grips with my RGB processing and, frankly, I've been tearing my hair out trying to get the processing right all day, but here's my best effort so far. I'm certain it could be improved upon, but I've had enough for the time being.

c. 36 x 300s Luminance, plus 12x300s each of RGB unbinned. SXVF-H9, 250mm f/4.7 Newtonian.

Here's another version, and I'd be grateful for any comment on which version you think is better.

[Jannis]

I haven't even seen this one before. looks nice

Tried to look it up in stellarium, and i found it, but no picture available there unfortunally.

I think i'd like something right in the middle of the two versjons you have there. But if i had to choose i'd say the 2nd one looks best to me. But i'm on an uncalibrated TV right now, so chances are i may change my mind if i go to my calibrated display and look again...

[Sam]

Nice work Luke - good to see a galaxy that's not imaged very often.

[foundaplanet]

Had it have been clear here last night I was going to try this myself so it is great to see it here now.

Good image, or two:). I like both ways you have processed it and it is interesting to look at the difference between the two.

Nice work.

[RobH]

Great job, and a good find....that's gone right to the top of the list!!

I reckon there's a lot more to be brought out in this image.

Your black is very black and you're losing faint detail, and probably quite a few faint fuzzies in the background too.

When you're processing, at all stages until the end, don't set your blackpoint any lower than about 20/20/20 for RGB (in photoshop).

This way, you can be sure you've not lost anything, and, if the background seems to light in the end, then darken it at that point, but I would suggest not going darker than 10/10/10 at the very darkest with a nebula or galaxy image.

Cheers

Rob

[lukebl]

Many thanks, gents, and thanks Rob for the advice. This image has been my biggest challenge in processing yet, revealing my weaknesses in Photoshop processing, and I just can't get it quite right. The capture was perfect for the first time: guiding, focus and collimation were spot-on, and I didn't have any image distortion through focus sag which I've had previously.

The biggest problem has been removing the uneven background gradient, despite using flats, which is why I've ended up with the backgound being too black. Here's a monochrome version, which in may ways I prefer although you can see the uneven gradient here. I've also had issues with getting the colour right. I did do 2 hours-worth of H-a, but that didn't seem to make any difference. Despite all that, it's been my most rewarding DSO so far, because I know that the raw data is as good as I've ever achieved, and it's all now down to the post-processing. Incidentally, Rob, I'm not sure what you mean about setting the blackpoint at 20/20/20. Can you explain, or direct me to a tutorial on this. Ta.

And here's a slightly reworked colour version. You see the uneven gradient.

[prokyon]

Thanks for sharing! Did not know it before. Great job!

[RobH]

Much better now

Do you have Gradient Xterminator? If not, it's well worth getting and will have no problems dealing with your gradient.

Have a look on my website under 'primers and tutorials'....there's a section there about image stretching etc.

In a nutshell though, in Photoshop, bring up LEVELS....in the dialogue box you'll see three eyedropper symbols....the left hand one is for your blackpoint. Double click this and a new box will appear that will allow you to set the RGB values....remember, zero for each of these is totally black. I do most of my processing with these set at 20 or 25.

Once your blackpoint is set in the dialogue box, simply click on an area of background in your image and these values will be applied.

HTH

Rob

[lukebl]

Thanks a lot Rob. I think I've used your tutorials before, but I'll have another look. I do have Gradient Xterminator. I found it very useful for my DSLR imaging, but with my CCD images it seems to make matters worse. I seem to end up with a halo of dark background around the DSO, fading to a paler noisy backround with increasing distance from it.

[RobH]

Are you selecting the background only Luke?

If not, GX can give some very odd results!

Rob

[lukebl]

Are you selecting the background only Luke?

If not, GX can give some very odd results!

Rob

Doh. It never occurred to me to use GX on just the backgound. I thought you had to do the whole image. Here's another version:

[Jannis]

Last processed one is absolutely the best looking if you ask me. Much more details and nice colours in it

[661-pete]

Very nice result there with lots of detail, in a little-known spiral - I'd never heard of it and it's not in Caldwell. I'm in two minds as to whether I prefer the version with the little bit of gradient, or the last one, can't decide which one is more pleasing to the eye. Lots of background faint fuzzies to look for, as well!

Thanks for introducing us to this!

[astro mick]

Very nice Luke.

TBH i,d be very happy with any of your images,but i think the last one has the edge.

I too struggle with Photo-Shop, especially after converting the RGB to 8bits in order to combine.I can usually tell nothing from the Histogramme,as it displays a lot of combing in 8 bits.(Dont know what the answer is to that).

I have had the mono images of M33 sat in my computer since the 2/11/11,and have as yet had no skies to get the RGB data.

I will add this galaxy to my list.

Thanks for posting.

Mick.

[lukebl]

Thanks again all. And Sam (lightbucket), just found your posting of an image of this very galaxy by your good self from a couple of years ago. It must be pretty low in the sky for you in NZ? Definitely an interesting subject, and strangely overlooked.

[RobH]

Looking much better now Luke.

There's still more to be brought out of the galaxy I reckon, with careful use of curves. You have enough good data there to be able to push it a little more before noise becomes an issue.

I doubt that with 5 minute 1x1 subs you've got many saturated stars, so you should be able to pull out some star colour too. I never use luminance layers on my stars if I can avoid doing so as it tends to wash the colour out very easily, and I usually do a second quick process just for the stars, then drop them in, or rub through the layer above to reveal them. You have to match up your black points on both layers, and feather the selections carefully, but it can work well.

Cheers

Rob

[lukebl]

Cheers again, guys. Here's a bit of a re-work. Just wish I could get to the bottom of the uneven gradients in the background. I think I now prefer the subtler effect of the earlier monochrome version. Anyway, it's been a worthwhile learning exercise.

[Starlight 1]

I like one before this one

the soft look for me.

[lukebl]

I like one before this one

the soft look for me.

Thanks. You're right, so do I!

[ollypenrice]

I agree with Rob's tips, which mirror my own method. By far the most powerful rectifier of backgrounds is Dynamic Background Extraction in PixInsight which would, I think, make light work of this. However, there is a tutorial on the Croman website explaining how to isolate the sky background. I rarely use GX these days and used to use Colour Select, myself, to pick the areas for Xtermination!

You have some dark star haloes, notably within the galaxy. Where do these come from? They look like sharpening artefacts from Deconvolution or unsharp mask etc. If sharpening I always select and then exclude the stars. You can correct haloes like this with a dab of Clone Stamp set to Blend Mode Lighten, but it is best not to create them in the first place of course.

Good to see star colour in the later ones but watch your blues, which are drifting into magenta.

Anyway, after all that, I'd like to join in the applause both for an excellent discovery and for a very fine image of it! Very fine indeed.

Olly

[Jessun]

Luke, I love the picture and how it's developing with the help of the guys!

This is stargazerlounge at its very best!

/Jessun

[lukebl]

....This is stargazerlounge at its very best!

/Jessun

You are so right! You NEVER stop learning with this lark/hobby/obsession, and there's always someone here to offer constructive advice and critcism, which is what keeps me going. Unfortunately, I now have about 20 versions of this galaxy cluttering up my hard drive. I think I am leaning towards some of my earlier, more subtle versions. Olly, I know my processing leaves a lot to be desired, but I think I'm getting there. I know what you mean about the haloes, but the red star in the lower part of the galaxy (ok, I know it's in our galaxy and not in NGC925!) actually does appear to occupy a darker part of it, as it looks like that with the minimum of stretching of curves and levels on the original TIF.

[Jessun]

Just remember that anyone touching those histogram sliders etc becomes an artist and you'll pretty much interpret the heavens. The HST pictures are also heavily edited - albeit by a master (using PS haha) - so not even those images provide us with the truth, or even an aim.

The photons travelled all that distance for you to play with! And thanks for sharing!

/Jessun

[Fay]

Well done Luke, a very nice galaxy you have brought to everyone's attention

Fay