This one was much harder to work with than I expected. It's actually quite faint, and there's a rich star field here which, when combined, makes it quite difficult to bring out the galaxy's details without overwhelming the image with stars.
Anyway, it's quite a nice FOV with lots of little galaxies floating around.
I may end up capturing more data on this, particularly Ha, but that will mean I'll have to process it again... sigh. Then again, I'll probably end up processing it 50 times anyway, and I still need to get rid of the weird colour hotpixels popping up around blown cores of the brighter stars.
A bigger view with gear/capture details is available on Astrobin.
PS: If anyone has any tips on multiscale processing in PI, they'd be greatly welcomed ;-)
Last edited by codemonkey; 08-08-2015 at 03:25 PM.
It is a really nice image, colours are well balanced and detail is very good. Maybe more Luminance (more subs) would help to make the galaxy stand out more?
Nice one. I attempted this one years ago from my little spot in suburbia with a 4" apo and my results were abysmal by comparison. It's a tough nut to crack but I think you've done well here. I agree more data in lum will help and a touch of ha blended in never hurts.
That's an excellent image Lee. Looks a little dark on my screen but there is quite a lot of detail visible, and I really like the colours. I can't help with the Ms processing as that is one PIS tool that I just cant figure out. Can'twait to see the final version if you end up taking more subs.
It is a really nice image, colours are well balanced and detail is very good. Maybe more Luminance (more subs) would help to make the galaxy stand out more?
Cheers mate! I think it was actually a bit too green. I'm not sure why, but after brightening it up a bit balancing the colours better (I think?) seems to give me a more red/magenta look than others I've seen. Maybe I've been looking at it too long and I've just stuffed the colour up though lol.
I've got a reasonable about of lum in this one (more always helps though) but the tricky part is not bloating the stars while trying to extract the galaxy. I think I just need to get better at processing more than extra data at this point.
Quote:
Originally Posted by AlexN
Nice one. I attempted this one years ago from my little spot in suburbia with a 4" apo and my results were abysmal by comparison. It's a tough nut to crack but I think you've done well here. I agree more data in lum will help and a touch of ha blended in never hurts.
Thanks Alex! Yeah it was definitely more challenging than I expected given the stats I saw on it before I tried it. I'm fortunate enough to have pretty dark skies here (bortle 3 I estimate). I tried one sub on a night with the moon up and very quickly shelved the idea of getting any more data without a totally moonless sky. I don't think I'd even try in suburban skies.
Quote:
Originally Posted by Rex
That's an excellent image Lee. Looks a little dark on my screen but there is quite a lot of detail visible, and I really like the colours. I can't help with the Ms processing as that is one PIS tool that I just cant figure out. Can'twait to see the final version if you end up taking more subs.
Thanks Rex! You're right, definitely dark. I've now given up on trying to keep the stars under control and reprocessed it to bring out the galaxy more so I think it's a bit better now... well, I hope so :-)
I'll update the OP with the new version in a sec, but it's on Astrobin already.
As Colin has already suggested, I would recommend trying applying some stretch with a star mask, or maybe cropping out stars, and adding them after the Galaxy has been properly stretched?
You did quite a nice job, Lee! It is a difficult target.
Might be worth trying MaskedStretch instead of HistogramTransform to control the stars a bit better. I don't use it a lot but some people like it.
You can also reduce the stars post stretch with a star mask and MorphologicalTransformation. Sometimes a star mask and a brightness reduction with CurvesTransform and/or HistogramTransform does a nice job too, especially for the largest stars.
It is looking pretty good at the moment, going to have a look at the high res version when I get home.
One suggestion for capturing the galaxy, you may be better off taking some deeper (longer) subs for the galaxy and cropping out the stars.
Cheers Colin :-) That could definitely work, but I'm not sure if that's crossed the line where it's become more effort than I'm prepared to put in lol.
Quote:
Originally Posted by Slawomir
As Colin has already suggested, I would recommend trying applying some stretch with a star mask, or maybe cropping out stars, and adding them after the Galaxy has been properly stretched?
Cheers S :-) I did play around with that idea briefly with this but found that it's a real PITA trying to get a mask that doesn't leave rings around at least some of the stars. Even when I focused on only doing large stars there'd still be some with rings while others were perfect.
Masked stretch did a better job of it than I could when trying this way, but that did leave stars looking like they have a bright core with a weird larger more diffuse halo.
Quote:
Originally Posted by RickS
You did quite a nice job, Lee! It is a difficult target.
Might be worth trying MaskedStretch instead of HistogramTransform to control the stars a bit better. I don't use it a lot but some people like it.
You can also reduce the stars post stretch with a star mask and MorphologicalTransformation. Sometimes a star mask and a brightness reduction with CurvesTransform and/or HistogramTransform does a nice job too, especially for the largest stars.
Cheers,
Rick.
Thanks Rick! :-) As above, I did try MS, with somewhat ok results. In the end I gave up on trying to keep the stars under control and did something I've never done before: I used STF to stretch the LRGB, put it all together and then tweaked it from there. Even though the stars are bigger I actually ended up with a better result when I didn't worry about blowing things out.
The thing with MT is that it mostly seems to make the stars dimmer rather than smaller, which can give the image a strange appearance if you're not careful with it.
Having said that, maybe a careful tweak with MT or curves/hist now that I pushed everything more will tidy it up a bit.
I really think I'm going to have to invest in a specialised astronomy processing software solution. Using DSS and Ps Cs6 is clearly lacking in a lot of this functionality you guys are discussing.
I really think I'm going to have to invest in a specialised astronomy processing software solution. Using DSS and Ps Cs6 is clearly lacking in a lot of this functionality you guys are discussing.
As much as I ***** about the UI in PixInsight, I think it's pretty hard to beat. Doesn't mean you can't get good results with DSS and PS though, just a different toolset to work with. You may even find that you're able to do better with PS (I doubt it, but hey, it's possible).
Quote:
Originally Posted by strongmanmike
I recon both versions look ok really but as is often the case with repro's, I find myself thinking..Hmm? maybe combine the two 50/50..?
Yes there are a few stars but hey, stars are "meant" to be there after all they are part of the scene
Mike
Cheers Mike :-)
I sometimes wonder if the initial version biases our view of subsequent versions. I think what I'll do is leave it for a week or so and have another look as I think I'll find it easier to see it then. Sounds weird, right? I just looked at my recent M83 again and it looks wrong to me.
The thing with MT is that it mostly seems to make the stars dimmer rather than smaller, which can give the image a strange appearance if you're not careful with it.
You can give them a little sharpen afterwards or stretch them a little with a tight star mask. Also worth trying a contour star mask to reduce just the edges of the stars with MT.
I was surprised how much Ha is in this galaxy. You've done very well Lee. I like the second rendition the most. I'm a warm nucleus kind of guy which the second image presents.
well done on a tricky target. i always find it is the smaller galaxies that you pick up that make an image incredibly interesting. you have managed to capture 6744's LMC look alike. it is also amazing how far the spiral arms extend, especially with the interaction of its companion galaxy. looks like a candidate for an extreme deep field
I was surprised how much Ha is in this galaxy. You've done very well Lee. I like the second rendition the most. I'm a warm nucleus kind of guy which the second image presents.
Thanks Jase! I did want to capture some Ha on this one, but then I discovered the little group containing 6770 and my attention was quickly diverted.
The colour balance on the first seems wrong, even though the latter image doesn't seem to look like the others I've seen. Not sure why that is, maybe the balance is somewhere in between.
Quote:
Originally Posted by Somnium
well done on a tricky target. i always find it is the smaller galaxies that you pick up that make an image incredibly interesting. you have managed to capture 6744's LMC look alike. it is also amazing how far the spiral arms extend, especially with the interaction of its companion galaxy. looks like a candidate for an extreme deep field
Thanks Aidan! This is definitely a candidate for a deep field, there's some cool stuff going on. To do it well I think you need better skies/skill than me to keep all those stars nice and small so that you can push it more to bring out those details.
Lee the acquisition looks pretty good. I think the star colour is nice too. The very faint outer arms are blue as you would expect and the bar look to be golden and correct, but I am not a fan of the magenta looking inner spiral arms. So I think the image is almost there.
Thanks Paul, appreciate it. I'll have another crack at the colour; you're right, it doesn't seem correct, as no other image I've seen of this has that colour. Not sure where I went wrong with the calibration though... anyway, thanks for the feedback!
they are part of the scene
