M31 HaLRGB

[Rodd]

No, not another reprocess....a whole new data set.  Since I lost all my old data, I needed new subs. I can't say I am overjoyed.  The data set should be good--3 hours each of RGB, over 4 hours of Lum and 3 hours of Ha.  That's over 3 times as much data as my last attempt with 1min subs instead of 30 sec subs--so it should be deeper.  .  But try as I might...….and I mightily tried, I found it very difficult to achieve satisfactory results.  Certainly not very much different than the old image--a bit deeper, a tad sharper, a little cleaner.  But the palette remains very similar.  I was able to improve M32 at least 

FSQ 106 with .6x reducer and ASI 1600.  

Red: 192 60 sec

Green: 18860 sec

Blue 165 60 sec

Lum 230 60 sec

Ha: 38 5min

If anybody is interested, I would very much like to post the full data set for those of you who want to try their hand at the data.  If I get a few yays I will post the fits files.  One thing is for certain.....I am sick of M31!

 

[Miguel1983]

Amazing colours !

I'm certainly no expert and i don't want to offend you, but i get the feeling the sharpening is a bit too much ?

Still, a great image !

[wimvb]

Looks mighty good to me, Rodd.

If you post the data, I'd love to have a go at it. I'm clouded in for the foreseeable future, and my processing fingers are itching like mad.

[wornish]

Fantastic image as always.

 

Please share the raw data files I would love to have a go at processing this amount of data. 

[coatesg]

Colour looks great to me too, and doesn't look overly crunchy, but there's some odd artifacts (dark ringing) in the halos of the bright stars.

I'd like to have a go at the data as well if you can post it!

[Laurin Dave]

Great image Rodd.. I’ll have a go too ..  be a lot of data though or are you just posting the LRGBHa stacks? 

Dave

[Swoop1]

Rodd- a lovely image!

Re your 'dissapointment' over the lack of marked improvement compared to previous data, perhaps you have pushed it to the limits and your image is perfect?

[carastro]

Fantastic image Rodd.  666 subs?  the mind boggles, I would also love to have a "go" at the processing and agree with others that on the zoomed in view there does seem a slight oversharpening which has affected the edges of the stars, but otherwise a superb bit of work.

Can I ask you your method of adding the Ha to the target which you have also done excellently as this is something I struggle with.  I have tried a number of methods but none of them produce results like this.

Thanks

Carole 

 

[souls33k3r]

Good God buddy, you're being over critical about it mate. It's a fantastic image. But knowing you, you won't rest till it's perfect in your eyes so here we are all rooting for you mate

[gorann]

4 hours ago, wimvb said:

Looks mighty good to me, Rodd.

If you post the data, I'd love to have a go at it. I'm clouded in for the foreseeable future, and my processing fingers are itching like mad.

Beautiful image Rodd! I am in the same situation as Wim and thanks to his comment I now know why my fingers are so itchy.....

[Sunshine]

Wow that’s a wonderful Andromeda! Ha really kicks it up a notch, beautiful colors and definition.

[Craney]

Cracking image.   

Your image really does give the impression of  gas spiralling into the galactic core.    A lot of M31 images have this aspect 'blown out' with over exposure.

 

[Rodd]

8 hours ago, Miguel1983 said:

Amazing colours !

I'm certainly no expert and i don't want to offend you, but i get the feeling the sharpening is a bit too much ?

Still, a great image !

Thanks Miguel.  I don't think the sharpening is way over the top--the data can support it,  There are no real sharpening artifacts--maybe a very tiny bit over.  Most of teh artifacts that can be seen ( the rings in teh halos, and things like that) are due to my all thumbs approach at processing!   I like high-resolution images. 

Rodd 

[Rodd]

7 hours ago, wimvb said:

Looks mighty good to me, Rodd.

If you post the data, I'd love to have a go at it. I'm clouded in for the foreseeable future, and my processing fingers are itching like mad.

Thanks Wim--data to come...I am saving teh files as FITs--give me a few minutes

Rodd

[Rodd]

5 hours ago, wornish said:

Fantastic image as always.

 

Please share the raw data files I would love to have a go at processing this amount of data. 

Thanks Wornish--files coming in a few.

Rodd

[Rodd]

5 hours ago, coatesg said:

Colour looks great to me too, and doesn't look overly crunchy, but there's some odd artifacts (dark ringing) in the halos of the bright stars.

I'd like to have a go at the data as well if you can post it!

AHHHHHH!!!!!  Darn it.  I had not seen that.  What a mess.  I have helped in along (I will post the images with the FITs files in a few).  I clone stamped it out but then decided that that was cheating.

Rodd

[Rodd]

5 hours ago, Laurin Dave said:

Great image Rodd.. I’ll have a go too ..  be a lot of data though or are you just posting the LRGBHa stacks? 

Dave

Thanks Dave...I will post the HaLRGB stacks--certainly not all the subs, man! I am glad I didn't have vthat thought when I was driving?

Rodd

[Rodd]

3 hours ago, Swoop1 said:

Rodd- a lovely image!

Re your 'dissapointment' over the lack of marked improvement compared to previous data, perhaps you have pushed it to the limits and your image is perfect?

Thanks Swoop.  As much as I would like to believe that was the case, I am not so foolish.......I hope to prove that with subsequent versions!

Rodd

[Rodd]

3 hours ago, carastro said:

Fantastic image Rodd.  666 subs?  the mind boggles, I would also love to have a "go" at the processing and agree with others that on the zoomed in view there does seem a slight oversharpening which has affected the edges of the stars, bit otherwise a superb bit of work.

Can I ask you your method of adding the Ha to the target which you have also done excellently as this is something I struggle with.  I have tried a number of methods but none of them produce results like this.

Thanks

Carole 

 

2

Thanks, Carole.  remember, this is CMOS and only 1 min subs, so the total time is not that impressive--the number of subs though is very inconvenient.  Filled up my hard drive with the registration subs and calibration subs.  I really need to replace my external drive.

regarding adding Ha.  I use PI--can't recall if you use PS or PI.  Anyway.  I remove the red continuum signal from the Ha stack (the red broadband leakage) using Pixel Math.  You extract the red channel from the RGB image and call it r.  Then you call the Ha stack h.  using Pixel Math with the "use 1 channel" option checked. you input the following formula into the red channel line (first line) and execute the linear state.

$T+(r-Med(r))*0.2.           the 0.2 is variable depending on the image it can range from .005 to .03 or so.  This will remove all signal from the stack except true Ha.  Then you clone the stack and turn the clone image nonlinear so it can be used as a mask on the linear h stack.  The idea is to make a well-fitting mask that is very dark where there is no Ha signal--black if possible.  You alternate with using curves to reduce the background, invert the mask and increase the Ha signal so the mask is bright where there is Ha and black where there is background.  You then use this mask on the RGB image and using Pixel Math, you uncheck the box for "use single channel) and use the following formula

$T+(h-Med(h))*3......Note the multiplier is much higher.  This is judgment.  It ranges from 1.5 to 4 or so.  Too high will result in too red and bright Ha regions after processing.  Too low and the Ha regions will be too pale.  This gets the Ha into the red channel just where there is Ha signal.

Now--you have to do the same thing to the linear Luminance stack (because there is a small amount of Ha signal in that too).  You basically do the same thing, use the same mask, and the same formula ($T+(h-Med(h))*.5....But notice the multiplier is much smaller.  Again judgment, but it is usually less than 1 as the Ha signal in the Lu stack is small.  

This gest the Ha in the Luminance stack (if you do it zoomed in you should just be able to see the addition--not a lot--just a bit).

Now--you stretch the Lum and the RGB so they have identical histograms (the linear fit tool can help---you use it on an extracted lightness from the RGB with the Lum stack as the reference, reinsert the extracted lightness, then use LRGB to insert Lum stack).  Or- you use your method of adding the lum to the RGB as normal.

I hope this helps (and I hope you use PI!).  If you don't use PI I am afraid it will be gibberish.

Rodd

[Rodd]

3 hours ago, souls33k3r said:

Good God buddy, you're being over critical about it mate. It's a fantastic image. But knowing you, you won't rest till it's perfect in your eyes so here we are all rooting for you mate

Thanks barking Monkey (couldn't resist).  I am afraid you are right!

Rodd

[Rodd]

3 hours ago, gorann said:

Beautiful image Rodd! I am in the same situation as Wim and thanks to his comment I now know why my fingers are so itchy.....

Coming soon

[Rodd]

1 hour ago, Craney said:

Cracking image.   

Your image really does give the impression of  gas spiralling into the galactic core.    A lot of M31 images have this aspect 'blown out' with over exposure.

 

Thanks Craney--I agree.  In fact many images lose contrast due to being overly bright.  M51 is another one.  I have reduced the brightness even more--See attached image. below

Rodd

[Rodd]

8 hours ago, Miguel1983 said:

Amazing colours !

I'm certainly no expert and i don't want to offend you, but i get the feeling the sharpening is a bit too much ?

Still, a great image !

 

7 hours ago, wimvb said:

Looks mighty good to me, Rodd.

If you post the data, I'd love to have a go at it. I'm clouded in for the foreseeable future, and my processing fingers are itching like mad.

 

6 hours ago, wornish said:

Fantastic image as always.

 

Please share the raw data files I would love to have a go at processing this amount of data. 

 

5 hours ago, coatesg said:

Colour looks great to me too, and doesn't look overly crunchy, but there's some odd artifacts (dark ringing) in the halos of the bright stars.

I'd like to have a go at the data as well if you can post it!

 

5 hours ago, Laurin Dave said:

Great image Rodd.. I’ll have a go too ..  be a lot of data though or are you just posting the LRGBHa stacks? 

Dave

 

4 hours ago, carastro said:

Fantastic image Rodd.  666 subs?  the mind boggles, I would also love to have a "go" at the processing and agree with others that on the zoomed in view there does seem a slight oversharpening which has affected the edges of the stars, bit otherwise a superb bit of work.

Can I ask you your method of adding the Ha to the target which you have also done excellently as this is something I struggle with.  I have tried a number of methods but none of them produce results like this.

Thanks

Carole 

 

 

3 hours ago, souls33k3r said:

Good God buddy, you're being over critical about it mate. It's a fantastic image. But knowing you, you won't rest till it's perfect in your eyes so here we are all rooting for you mate

 

3 hours ago, gorann said:

Beautiful image Rodd! I am in the same situation as Wim and thanks to his comment I now know why my fingers are so itchy.....

 

1 hour ago, Sunshine said:

Wow that’s a wonderful Andromeda! Ha really kicks it up a notch, beautiful colors and definition.

 

1 hour ago, Craney said:

Cracking image.   

Your image really does give the impression of  gas spiralling into the galactic core.    A lot of M31 images have this aspect 'blown out' with over exposure.

 

Thanks all--Here are the FITs stacks.  I have cropped them and removed the gradients (there were not many--no Moon and most data shot between 60 and 90 degrees).  Otherwise, these are raw stacks. I have posted a revision as well--I could not completely fix the rings in the aura of the bright star at right.  That will require a reprocess (god help me).  The data looks good to me--though FWHM values may be a bit high.  The initial RGB combined image looks amazingly deep and detailed (low on contrast though as usual), with real looking stars and a sense of structure in the spiral arms.  I feel that I have degraded the image since the channel combination.  All of my processing steps resulted in incremental degradation from an initial state of clarity.  I really need to learn how to process the darned data!

R192-DC-DBE.fit

G188-DC-DBE.fit

B186-DC-DBE.fit

L231-DC-DBE.fit

Ha-38 DC-DBE.fit

 

My new image (just a small tweak)

[RolandKol]

43 minutes ago, Rodd said:

Thanks, Carole.  remember, this is CMOS and only 1 min subs, so the total time is not that impressive--the number of subs though is very inconvenient.  Filled up my hard drive with the registration subs and calibration subs.  I really need to replace my external drive.

regarding adding Ha.  I use PI--can't recall if you use PS or PI.  Anyway.  I remove the red continuum signal from the Ha stack (the red broadband leakage) using Pixel Math.  You extract the red channel from the RGB image and call it r.  Then you call the Ha stack h.  using Pixel Math with the "use 1 channel" option checked. you input the following formula into the red channel line (first line) and execute the linear state.

$T+(r-Med(r))*0.2.           

.............

.................

Now--you stretch the Lum and the RGB so they have identical histograms (the linear fit tool can help---you use it on an extracted lightness from the RGB with the Lum stack as the reference, reinsert the extracted lightness, then use LRGB to insert Lum stack).  Or- you use your method of adding the lum to the RGB as normal.

I hope this helps (and I hope you use PI!).  If you don't use PI I am afraid it will be gibberish.

Rodd

WOW!
I been waiting for your stacks, to experiment with HaRGB  as I never done it before
but reading this, -  just completely confused me
It sound more like a wizardry and witchcraft!  but it produces a VERY nice result!

I will need to learn this "abracadabra" before I buy my LRGB filters!
Thanks for stacks!

 

[Rodd]

6 minutes ago, RolandKol said:

It sound more like a wizardry and witchcraft!

How else to keep the Gremlins at bay? Seriously, it can be a bit overwhelming, and I always have to look at the cheat sheet to get the formula right.  In fact, I lost the cheat sheet recently and hope I got them right here!.  Fortunately,  There is a tutorial online.  Good luck!

Rodd