Leo Triplets LRGB

[Rodd]

26 minutes ago, kirkster501 said:

I love it.

Thanks kirkster

Rodd

[gorann]

The best of luck to you tonight Rodd - you are on to a winner!

[wimvb]

22 hours ago, Rodd said:

Thanks Olly--you know its strange but this is not the first time a "short" channel dominated the color--it tends to happen for me for some reason.

Rodd 

This is to be expected if one channel dominates noise-wise, and the peaks of the histograms are aligned. The "short" channel has a wider bell curve (histogram), so more bright pixels, or higher level to the right of the peak. An easy way to remedy this (in PI) is to use scnr on that channel, but at less than 100 %. In PS you'd probably just use colour sliders to make the background look neutral. That also works in PI.

(Best way to remedy this is of course to collect more data.)

[Rodd]

1 hour ago, wimvb said:

This is to be expected if one channel dominates noise-wise, and the peaks of the histograms are aligned. The "short" channel has a wider bell curve (histogram), so more bright pixels, or higher level to the right of the peak. An easy way to remedy this (in PI) is to use scnr on that channel, but at less than 100 %. In PS you'd probably just use colour sliders to make the background look neutral. That also works in PI.

(Best way to remedy this is of course to collect more data.)

Ah.  That explains why I saw such an impact when I used scnr to get rid of what appeared to be a pervasive blue cast, which surprised me considering the number of blu subs

Rodd

[ollypenrice]

When I did the Triplet, Rodd, I used two layers in order to get the background quiet. We all need a manic stretch to get the tail and the outer galaxy details and this tends to bring up sky noise. So I did a gentle stretch just for the background and stars, taking great care to get the background sky values identical in this soft stretch to the brightness of the sky in tha manic stretch. However, although the final values were the same the soft stretch background was much quieter and the stars smaller. I pasted the soft stretch on top in Ps, rendered it zero opacity but active, then ran the eraser over it wherever I could find galaxy, tidal effects or faint fuzzy (of which there are lots.) When you bring the opacity back up on the top layer you have the best of both stretches. To my mind this beats masking.

Olly

[Rodd]

21 minutes ago, ollypenrice said:

When I did the Triplet, Rodd, I used two layers in order to get the background quiet. We all need a manic stretch to get the tail and the outer galaxy details and this tends to bring up sky noise. So I did a gentle stretch just for the background and stars, taking great care to get the background sky values identical in this soft stretch to the brightness of the sky in tha manic stretch. However, although the final values were the same the soft stretch background was much quieter and the stars smaller. I pasted the soft stretch on top in Ps, rendered it zero opacity but active, then ran the eraser over it wherever I could find galaxy, tidal effects or faint fuzzy (of which there are lots.) When you bring the opacity back up on the top layer you have the best of both stretches. To my mind this beats masking.

Olly

No doubt you are correct Olly--but alas, the Lone Ranger is stuck with Tonto.  Great idea regarding the 2 versions, I was thinking along these lines last night.  I can use Pixel Math to replace the background of the noisier image with the background of the smoother image (once I have them where I want them).  Or, conversely, replace the galaxies in the quieter image with the Galaxies in the nosier image.  

Rodd

[ollypenrice]

1 hour ago, Rodd said:

No doubt you are correct Olly--but alas, the Lone Ranger is stuck with Tonto.  Great idea regarding the 2 versions, I was thinking along these lines last night.  I can use Pixel Math to replace the background of the noisier image with the background of the smoother image (once I have them where I want them).  Or, conversely, replace the galaxies in the quieter image with the Galaxies in the nosier image.  

Rodd

It's a while since I thought about the Lone Ranger!

Olly

[wimvb]

3 hours ago, Rodd said:

I can use Pixel Math to replace the background of the noisier image with the background of the smoother image

Pixel Math & Clone Stamp to the rescue. But if you want to mess it up, Rogelio Bernal Andreo described a much more elaborate way in Lessons From the Masters.

http://www.deepskycolors.com/archivo/2010/05/07/multi-scale-Processing--Revealing-very.html

[Rodd]

1 hour ago, wimvb said:

Pixel Math & Clone Stamp to the rescue. But if you want to mess it up, Rogelio Bernal Andreo described a much more elaborate way in Lessons From the Masters.

http://www.deepskycolors.com/archivo/2010/05/07/multi-scale-Processing--Revealing-very.html

Wow-thanks Wim....I started giving it a quick read and soon realized I will have to give it a thorough read through.  In the case of the background the goal is the opposite--to eliminate.  But perhaps there are details in the galaxies  that can be brought forth using this technique--and certainly in nebula images.  

Rodd 

[wimvb]

You need a flat background for this to work. So, you'll need to spend time on dbe. It can help to use an overstretched copy of your image to place dbe samples. Then apply to the real image.

Even if this technique isn't as efficient as Olly's PS approach, you'll learn a lot of new tricks along the way.

[Rodd]

5 minutes ago, wimvb said:

You need a flat background for this to work. So, you'll need to spend time on dbe. It can help to use an overstretched copy of your image to place dbe samples. Then apply to the real image.

Even if this technique isn't as efficient as Olly's PS approach, you'll learn a lot of new tricks along the way.

Yes, it sounds like a well worth plumbing.  Thanks!

Rodd

[johnrt]

6 hours ago, ollypenrice said:

When I did the Triplet, Rodd, I used two layers in order to get the background quiet. We all need a manic stretch to get the tail and the outer galaxy details and this tends to bring up sky noise. So I did a gentle stretch just for the background and stars, taking great care to get the background sky values identical in this soft stretch to the brightness of the sky in tha manic stretch. However, although the final values were the same the soft stretch background was much quieter and the stars smaller. I pasted the soft stretch on top in Ps, rendered it zero opacity but active, then ran the eraser over it wherever I could find galaxy, tidal effects or faint fuzzy (of which there are lots.) When you bring the opacity back up on the top layer you have the best of both stretches. To my mind this beats masking.

Olly

Hee hee, you'll have the Pixinsight data purists choking on their tiffin with talk like this!

[carastro]

Stunning

[Rodd]

1 minute ago, carastro said:

Stunning

Thanks Carole!

Rodd

[ollypenrice]

40 minutes ago, johnrt said:

Hee hee, you'll have the Pixinsight data purists choking on their tiffin with talk like this!

I like to think so! There is something of the Spanish Inquisition about them...

lly 

[Rodd]

1 hour ago, ollypenrice said:

I like to think so! There is something of the Spanish Inquisition about them...

lly 

See below.

2 hours ago, wimvb said:

You need a flat background for this to work. So, you'll need to spend time on dbe. It can help to use an overstretched copy of your image to place dbe samples. Then apply to the real image.

Even if this technique isn't as efficient as Olly's PS approach, you'll learn a lot of new tricks along the way.

Well--all kidding aside (or all seriousness aside), The Ha idea is not looking too fruitful.  I collected 4 hours of Ha last night-8 30 min subs.  There is appreciable Ha signal in all 3 galaxies, but combining it with the red and lum channels does not have the same effect as say M33--where isolated star forming regions were prominent.  Perhaps the scale is too small to allow this type of regional distinction.  To get any visible effect I had to resort to inserting the Ha into the above, fully processed JPEG image.  Definitely not the best way to do it.  Even with this the effect is a slight reddening of the discs, with no knots of HII regions visible.    Maybe I am tired.  I will try again tomorrow.  

 

[johnrt]

3 hours ago, ollypenrice said:

I like to think so! There is something of the Spanish Inquisition about them...

lly 

[Rodd]

1 hour ago, johnrt said:

Now we’re just being silly?

[wimvb]

10 hours ago, johnrt said:

John, you might want to use DBE on this clip. There's a horrible gradient top blue to bottom red. 

[Rodd]

On ‎1‎/‎27‎/‎2018 at 10:02, ollypenrice said:

It's a while since I thought about the Lone Ranger!

Olly

see below

8 hours ago, wimvb said:

John, you might want to use DBE on this clip. There's a horrible gradient top blue to bottom red. 

see below

 

 

Well,  I was able to get the Ha to the targeted areas--but the scale makes it hard to be precise (at least in PI for me).  I realize the saturation is a bit high, and imbalanced (the blue is probably too much).  But as an exercise to show that there are indeed star forming regions in the hamburger and M66, I suppose it does succeed in telling us what we already knew!  I think withy better data there would be substantial improvement (the old Televue data is not the best).  I did use the method of using 2 images , 1 for background and stars and 1 for the galaxies--but I think I made the switch to early as there were considerable histogram modifications after that point.  The background is, therefore, was not the best. So, I used the background of the previous non Ha inclusive image and the new Ha enhanced galaxies.  Almost there!

 

 

[Rodd]

I managed to tone down the Ha and the blue--cheating really--I combined 2 images--the last and the second to last at a 60/40 ratio.  I think I will call this one done.

[ollypenrice]

This is the first time I've seen Ha added to the glorious Triplet and it tells me that I want to give it a go!

Olly

[Rodd]

2 minutes ago, ollypenrice said:

This is the first time I've seen Ha added to the glorious Triplet and it tells me that I want to give it a go!

Thanks Olly--I am sure you will produce a great image.  

Rodd 

[Rodd]

I could not tresist 1 more post--better background.  Lost a touch of tail perhaps--but it seems cleaner

[Rodd]

Some final fine scale sharpening.