new pic - Markarian's Chain

[glowingturnip]

Hi all,  here's my latest pic, Markarian's chain in Virgo:

20ish x300s lights, ISO800, darks, flats and bias, equipment as per sig, DSLR at prime focus, Pixinsight processing

Not my greatest, since it was 3 days after full moon and the moon was up, about 60 degrees away at the time, but beggars can't be choosers. 

Anyhow, hope you enjoy

[Leveye]

This is with your Celestron?

[glowingturnip]

yep

[michael.h.f.wilkinson]

Very nice!

[GuillermoBarrancos]

Impressive! To be proud of with the equipment you used!

[glowingturnip]

 and what's wrong my my equipment ?!? 

 (kidding )

[Leveye]

Nice! Great group caught here sir.

[laser_jock99]

Nice one.

[DaveS]

If you can do that from London then I'll have to pull my socks up ang get on with imaging instead of whingeing.

[Astrosurf]

Nice one!

Alexxx

[Dragon Man]

Fantastic work Stuart!

Look at all those tiny background galaxies 

[glowingturnip]

If you can do that from London then I'll have to pull my socks up ang get on with imaging instead of whingeing.

no no, have to confess, it's from a dark sky site in southern Spain

thanks guys !

[oldpink]

was wondering
I'm in London often and it has no stars unless you go way out from the city
though Epson isn't too Bad you see a few down there

[glowingturnip]

yeah, it's shocking isn't it, my house is next door to a floodlight sports pitch as well, lovely orange glow, and can barely even make out the brightest constellations on a clear night

[DaveS]

Oh, right, I did wonder why I couldn't match that

Floodlit pitches, driving ranges, and railway depots, Argh! . Why do people have to push asst balls around at night? Sports, pah!

Anyway enough of my ranting, that's a very nice galaxy photo, envious.

[glowingturnip]

cheers mate 

[glowingturnip]

re-did this one too, as I think I was guilty of the same over-aggressive noise reduction and stretching as with my M106.  I think this one's better in a more subtle way, and I was able to bring out more star colour too...

[Scott]

I love this chunk of sky and sadly, I fear I may have missed it for another year . 

Thanks for posting Stuart and I like the re-done version for what it's worth

[IanL]

I think this one's better in a more subtle way, and I was able to bring out more star colour too...

Are you using the masked stretch process in PI? Instead of histogram transformation when you are ready to do your first stretch, use masked stretch and you should get a lot more colour in the stars, etc. Then use the curves process and select saturation rather than the default RGB/K setting; make a nice big upwards curve from bottom left to top right and it will bring out the colour a lot more (you may want to use a mask to protect the background at this stage).

The only problem with masked stretch is that you will end up with less contrast/structure in the galaxies compared to a normal histogram transformation, but you can tweak this with other processes after the first stretch.  The trick is that the masked stretch doesn't blow out the highly saturated cores of stars and other features.  You look to have some good data in there and I reckon you can get more out of it without too much of a struggle.

[glowingturnip]

Hi Ian,

Yes, have been using a combination of those techniques - I normally split the calibrated raw into a synthetic lum and rgb which receive slightly different processing.  I do a histo stretch on the Lum but find that the masked stretch gives a nice bell-shape to the stars at the expense of worse contrast in galaxies etc, as you say, so I normally do both stretches, then do a star-masked opacity blend (with pixelmath) of the MS stars into the regular histo image.

You have a good point though, I just do a histo stretch of the RGB at the moment, I shall experiment with doing an MS stretch on the RGB instead, where the contrast is less of an issue and it might be kinder on the colours.

As you say, I use a masked curves tool on the saturation, and also add a slight s-shape to the a and b channels in the same tool.  Soft-light blending the RGB image with itself (pixelmath)  can separate out the colours nicely too, and little tweaks with the colour saturation tool.

(apologies to non Pixinsight users for the utter gobbledegook I've just written !)

[IanL]

Interesting idea for selective processing of stars vs. other features.  I tend to approach it the other way and just apply a masked stretch, then use a star mask to protect the stars and use LHE, HDR compression, etc. to give more contrast to the structures.  I don't split in to Lum and RGB for most of the processing since the tools where it really matters all tend to have settings that can target either or both separately within the same image.

I do sometimes split out the RGB towards the end and blur it if there is still remaining (lumpy) chroma noise before recombining, but it feels a bit too much like cheating so I try to avoid it if possible.

I reckon the masked stretch on the RGB may well be what you need, combined with a really aggressive curve (or even two) on saturation at the end of the process.

[glowingturnip]

cheating ?!  me ?!  never  ;-)

yep, will give the MS stretch on RGB a go, sounds like a good idea