Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Moderator: marktownley
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi all,
Although this is something discussed in the past, I could not find a satisfying answer on the best way to approach the real H-alpha colour.
I am aware that colour is a personal perception; we do not see it all in the same way, we usually do not have calibrated monitors etc. etc.
If something like the 'true H-alpha colour' exists (I hope it exists since it can be recorded by instruments and we can get measurements and absorption line in the spectrum), my practical question is what would be the best way to represent it when colourising a grayscale image of H-alpha.
For example, could we unanimously agree that playing with the RGB levels in photoshop, there are three values that produce 'true H-alpha colour'?
If such a thing does not exist objectively, then I would be happy to know what values you are using anyway in practice, when you wish to represent the H-alpha colour by RGB, or any other process you are using with an appropriate software.
Best wishes,
Alexandros
PS: There is an action in Astronomy Tools for photoshop to do what I am looking for, but the colour it gives seems far away from what we can see in H-alpha scope, I think. It is much more pink/violet than what I think is the H-alpha colour.
Although this is something discussed in the past, I could not find a satisfying answer on the best way to approach the real H-alpha colour.
I am aware that colour is a personal perception; we do not see it all in the same way, we usually do not have calibrated monitors etc. etc.
If something like the 'true H-alpha colour' exists (I hope it exists since it can be recorded by instruments and we can get measurements and absorption line in the spectrum), my practical question is what would be the best way to represent it when colourising a grayscale image of H-alpha.
For example, could we unanimously agree that playing with the RGB levels in photoshop, there are three values that produce 'true H-alpha colour'?
If such a thing does not exist objectively, then I would be happy to know what values you are using anyway in practice, when you wish to represent the H-alpha colour by RGB, or any other process you are using with an appropriate software.
Best wishes,
Alexandros
PS: There is an action in Astronomy Tools for photoshop to do what I am looking for, but the colour it gives seems far away from what we can see in H-alpha scope, I think. It is much more pink/violet than what I think is the H-alpha colour.
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
- MAURITS
- Way More Fun to Share It!!
- Posts: 7178
- Joined: Tue Nov 27, 2018 4:37 pm
- Location: Belgium
- Has thanked: 1721 times
- Been thanked: 3780 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Alexandros, I follow this post interested.
- Bob Yoesle
- Almost There...
- Posts: 958
- Joined: Thu Mar 08, 2012 7:24 pm
- Has thanked: 376 times
- Been thanked: 691 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Alexandros,
There certainly are a lot of RGB color conversion applications to achieve a realistic and valid color approximation. Here's one that is easy to use by just entering the wavelength and it gives you the corresponding RGB values.
My results for the Baader continuum filter at 540 nm:
CaK emission 394 nm:
H alpha emission 656 nm:
It couldn't be easier.
Bob
There certainly are a lot of RGB color conversion applications to achieve a realistic and valid color approximation. Here's one that is easy to use by just entering the wavelength and it gives you the corresponding RGB values.
My results for the Baader continuum filter at 540 nm:
CaK emission 394 nm:
H alpha emission 656 nm:
It couldn't be easier.
Bob
Diagonally parked in a parallel universe.
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
- rsfoto
- Way More Fun to Share It!!
- Posts: 5579
- Joined: Mon Jun 18, 2012 8:30 pm
- Location: San Luis Potosi, México
- Has thanked: 7350 times
- Been thanked: 4797 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Bob,
That is interesting. 656nm gives RGB 255,0,0
That is interesting. 656nm gives RGB 255,0,0
regards Rainer
Observatorio Real de 14
San Luis Potosi Mexico
North 22° West 101°
Observatorio Real de 14
San Luis Potosi Mexico
North 22° West 101°
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
This is really great!Bob Yoesle wrote: ↑Sun Dec 11, 2022 8:47 pm Hi Alexandros,
There certainly are a lot of RGB color conversion applications to achieve a realistic and valid color approximation. Here's one that is easy to use by just entering the wavelength and it gives you the corresponding RGB values.
It couldn't be easier.
Bob
Thanks so much Bob!
Two remarks:
1) H-alpha could not be more red.... it is RGB 255, 0, 0 :-)
2) I observed that this algorithm makes sudden passes of colour from some precise values to the next in nm, but then keeps the same colour (RGB) for a range of the spectrum. The full red 255, 0, 0 corresponds to all wavelengths from 645 to 700.
I found some on-line applications, but would like to make the conversion in PS. How did you convert the grayscale to H-alpha (255, 0, 0) monochrome?
And excellent example photos!!!! including the precious Venus transit that we won't see again during our lives, if I am not wrong... :-(
Thanks once more!
Best wishes,
Alexandros
Last edited by H-Alpha on Sun Dec 11, 2022 10:11 pm, edited 1 time in total.
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
-
- Way More Fun to Share It!!
- Posts: 6879
- Joined: Mon May 18, 2020 4:45 pm
- Location: Essex, S.E.England
- Been thanked: 4910 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Of course there are also PaintShopPro programs etc, that may well do the job, as well as using Ha scopes and cameras to achieve the desired results..
Terry
Terry
- Bob Yoesle
- Almost There...
- Posts: 958
- Joined: Thu Mar 08, 2012 7:24 pm
- Has thanked: 376 times
- Been thanked: 691 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Alexandros,
I'm not familiar with PS, as I use Paint Shop Pro instead, but the process will be essentially the same.
Here's an example greyscale image:
In PSP I choose Colors > Split channel > Split to RGB. This will give you 3 identical greyscale R, G, and B images to manipulate.
Select the appropriate R, G, and B image and apply a change in Colors > Brightness/Contrast adjustment level to match the individual RGB brightness values (not contrast - leave that alone) to what they should be. You then will have your separate RGB adjusted images to recombine:
Next your recombine your new R, G, and B images - Colors > Combine channel > Combine from RGB. You'll then have your selected color value image:
Bob
I'm not familiar with PS, as I use Paint Shop Pro instead, but the process will be essentially the same.
Here's an example greyscale image:
In PSP I choose Colors > Split channel > Split to RGB. This will give you 3 identical greyscale R, G, and B images to manipulate.
Select the appropriate R, G, and B image and apply a change in Colors > Brightness/Contrast adjustment level to match the individual RGB brightness values (not contrast - leave that alone) to what they should be. You then will have your separate RGB adjusted images to recombine:
Next your recombine your new R, G, and B images - Colors > Combine channel > Combine from RGB. You'll then have your selected color value image:
Bob
Diagonally parked in a parallel universe.
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
- rigel123
- Way More Fun to Share It!!
- Posts: 4648
- Joined: Thu Apr 08, 2021 9:26 pm
- Location: Mason, OH
- Has thanked: 5071 times
- Been thanked: 5575 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
All you need to do in Photoshop is once you open your grayscale image go to Image/Mode and select RGB. You can then use a variety of methods to make the color mix you want.
Warren
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
- robert
- Way More Fun to Share It!!
- Posts: 2992
- Joined: Fri Oct 14, 2011 1:49 pm
- Location: N.W.Scotland
- Has thanked: 4 times
- Been thanked: 1042 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Alexandros
I guess "objectively" is the problem here. Bob's images (accurate as they are, within the limits of RGB screens) show that there is a considerable loss of clarity of detail and contrast by doing this. Why would you want to do it? If the James Webb space telescope images were shown "objectively" they would be completely black for the mid infra red camera images.
This is a quick example of red only channel RGB compared to my adjusted image from 10th Dec to illustrate the losses
I use the same adjustment, more or less, in photoshop for my images but don't expect anyone else to agree with my personal aesthetic choice.
I use the Mode and RGB setting as Warren mentioned, then in "image adjust" I "load" an .alv file which does most of the adjustment (B down, G down a bit, R as is) and then adjust the Red channel to make the background pure black if necessary. I also sometimes brighten the image.
It is far from "objective" but I do share your idea that it is good to be consistent and use a standard adjustment for my images.
(I will attach my ha2.alv file in case you want to test it out! I tried but actually I can't attach it, but I could PM you if you want)
Robert
PS I also alter the 16 bit to 8 bit (in Mode) so that the .jpg file is usable!
I guess "objectively" is the problem here. Bob's images (accurate as they are, within the limits of RGB screens) show that there is a considerable loss of clarity of detail and contrast by doing this. Why would you want to do it? If the James Webb space telescope images were shown "objectively" they would be completely black for the mid infra red camera images.
This is a quick example of red only channel RGB compared to my adjusted image from 10th Dec to illustrate the losses
I use the same adjustment, more or less, in photoshop for my images but don't expect anyone else to agree with my personal aesthetic choice.
I use the Mode and RGB setting as Warren mentioned, then in "image adjust" I "load" an .alv file which does most of the adjustment (B down, G down a bit, R as is) and then adjust the Red channel to make the background pure black if necessary. I also sometimes brighten the image.
It is far from "objective" but I do share your idea that it is good to be consistent and use a standard adjustment for my images.
(I will attach my ha2.alv file in case you want to test it out! I tried but actually I can't attach it, but I could PM you if you want)
Robert
PS I also alter the 16 bit to 8 bit (in Mode) so that the .jpg file is usable!
Last edited by robert on Mon Dec 12, 2022 2:47 pm, edited 7 times in total.
images and animations http://tinyurl.com/h5bgoso
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Warren and thanks for this.
I tried the "Photo Filter" under "Image/Adjustments", but did not find it satisfying.
Best wishes,
Alexandros
Can you please indicate one of the methods you use? Just to clarify that my goal is to apply the 255, 0, 0 on the grayscale photo and not to play with Levels, Curves, Selective colour, Colour balance etc. in order to achieve the colour that I like the best.
I tried the "Photo Filter" under "Image/Adjustments", but did not find it satisfying.
Best wishes,
Alexandros
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
- robert
- Way More Fun to Share It!!
- Posts: 2992
- Joined: Fri Oct 14, 2011 1:49 pm
- Location: N.W.Scotland
- Has thanked: 4 times
- Been thanked: 1042 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
"my goal is to apply the 255, 0, 0 on the grayscale photo"
If you want to do this in PS you can convert to RGB and then select the "Channels" tab in the "Layers" window and go to G and B channels and fill them with black, then select RGB and have a look. This is how I did my quick example above.
Robert
If you want to do this in PS you can convert to RGB and then select the "Channels" tab in the "Layers" window and go to G and B channels and fill them with black, then select RGB and have a look. This is how I did my quick example above.
Robert
images and animations http://tinyurl.com/h5bgoso
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
- MAURITS
- Way More Fun to Share It!!
- Posts: 7178
- Joined: Tue Nov 27, 2018 4:37 pm
- Location: Belgium
- Has thanked: 1721 times
- Been thanked: 3780 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Thanks Bob to pointing out the colors.
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Thanks once more Robert for all your help and the PM regarding the .alv file! :-)
Once more our posts were crossed.
It may be applied only for only for pure RED (255, 0, 0) or pure G and B, but only for these precise monochrome situations. If you have any other idea for all monochrome colours is welcome.
For H-alpha, combining the website proposed by Bob's with your idea on RGB Channels we have the solution! :-)
Best wishes,
Alexandros
Once more our posts were crossed.
Yes, your idea is the solution for H-alpha. :-)robert wrote: ↑Mon Dec 12, 2022 2:51 pm "my goal is to apply the 255, 0, 0 on the grayscale photo"
If you want to do this in PS you can convert to RGB and then select the "Channels" tab in the "Layers" window and go to G and B channels and fill them with black, then select RGB and have a look. This is how I did my quick example above.
Robert
It may be applied only for only for pure RED (255, 0, 0) or pure G and B, but only for these precise monochrome situations. If you have any other idea for all monochrome colours is welcome.
For H-alpha, combining the website proposed by Bob's with your idea on RGB Channels we have the solution! :-)
Best wishes,
Alexandros
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Thanks a lot Bob!Bob Yoesle wrote: ↑Mon Dec 12, 2022 2:40 am Hi Alexandros,
I'm not familiar with PS, as I use Paint Shop Pro instead, but the process will be essentially the same.
Here's an example greyscale image:
Bob
Very intersecting, easy and useful menu in Paint Shop Pro! Good to know if I cannot find the equivalent action/menu in PS.
Thank you all for being so helpful!
Best wishes,
Alexandros
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
- robert
- Way More Fun to Share It!!
- Posts: 2992
- Joined: Fri Oct 14, 2011 1:49 pm
- Location: N.W.Scotland
- Has thanked: 4 times
- Been thanked: 1042 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Alexandros
I use CaK .alv file but if I tried to do an "objectively" coloured image to represent the eyepiece view it would have to be all black. My eyes don't see it at all....
Good luck
Robert
I use CaK .alv file but if I tried to do an "objectively" coloured image to represent the eyepiece view it would have to be all black. My eyes don't see it at all....
Good luck
Robert
images and animations http://tinyurl.com/h5bgoso
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
- rigel123
- Way More Fun to Share It!!
- Posts: 4648
- Joined: Thu Apr 08, 2021 9:26 pm
- Location: Mason, OH
- Has thanked: 5071 times
- Been thanked: 5575 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Yep, Robert gave you one of the more straight forward methods to get to a color you like. As you mentioned, you can also use an adjustment layer and play with levels on all three channels (RGB). It is a challenge to get that deep red color that you see visually and still see all of the nice details when working with an image. One thing I just thought of but haven't tried it yet is to make a deep red disk and then use a grayscale image as a layer to use as a Luminance layer like in deep sky processing.H-Alpha wrote: ↑Mon Dec 12, 2022 1:05 pm Hi Warren and thanks for this.
Can you please indicate one of the methods you use? Just to clarify that my goal is to apply the 255, 0, 0 on the grayscale photo and not to play with Levels, Curves, Selective colour, Colour balance etc. in order to achieve the colour that I like the best.
I tried the "Photo Filter" under "Image/Adjustments", but did not find it satisfying.
Best wishes,
Alexandros
I'll try posting this one more time, Chat has been really finicky today trying to get a post to actually post! Not sure what is up. I tried a couple of ways in PS to colorize a full disk I took on 12/4. This was the best method I found simply changing the mode to RGB and then used Color Balance and slid the Red slider all the way to the right and the G and B all the way to the left. Not too bad but I still see a loss of detail like I do anytime I colorize an image. Guess I'm a grayscale kind of guy!
Grayscale Colorized
Warren
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
- Bob Yoesle
- Almost There...
- Posts: 958
- Joined: Thu Mar 08, 2012 7:24 pm
- Has thanked: 376 times
- Been thanked: 691 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Warren & Robert,
With my RGB processing, I find it works best when you don't use jpg or other significantly compressed file formats.
Here's a full disc grey scale image that I had saved as a png, RGB split and combined for an Ha colored png. I then copied both, reduced the size, and then saved the comparison as a smaller jpeg. Loss of detail in the red channel image with jpg compression becomes obvious.
Open in a new tab to see a larger image.
However, here's crops of the original png full disc images grey scale and RGB split/combine. There appears there's no detectable degradation or loss of detail in the red channel png vs grey scale - at least as far as I can observe.
Bob
With my RGB processing, I find it works best when you don't use jpg or other significantly compressed file formats.
Here's a full disc grey scale image that I had saved as a png, RGB split and combined for an Ha colored png. I then copied both, reduced the size, and then saved the comparison as a smaller jpeg. Loss of detail in the red channel image with jpg compression becomes obvious.
Open in a new tab to see a larger image.
However, here's crops of the original png full disc images grey scale and RGB split/combine. There appears there's no detectable degradation or loss of detail in the red channel png vs grey scale - at least as far as I can observe.
Bob
Last edited by Bob Yoesle on Tue Dec 13, 2022 3:24 pm, edited 1 time in total.
Diagonally parked in a parallel universe.
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
Curiosity is the father of knowledge; uncertainty is the mother of wisdom.
Dark-Sky Defenders
Goldendale Observatory
- rigel123
- Way More Fun to Share It!!
- Posts: 4648
- Joined: Thu Apr 08, 2021 9:26 pm
- Location: Mason, OH
- Has thanked: 5071 times
- Been thanked: 5575 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
I think it comes down to my eyes. In both examples the colorized version is darker so I see better details in the grayscale. Interesting to see different methods and their results.
Warren
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
- H-Alpha
- Almost There...
- Posts: 579
- Joined: Sun Feb 21, 2021 2:19 am
- Location: Vouliagmeni (South Athens suburbs), Greece
- Has thanked: 1654 times
- Been thanked: 955 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Thanks Warren,
Anyway, as you say, I also have the feeling that the Red we simulate in the computer looses quite some details in comparison to grayscale and also in comparison to other false colour (like the usual golden/yellow/orange many of us use for both H-alpha and white light. Even in Bob's cropped example, my eyes perceive less detail/contrast that in his grayscale (or is it just my personal impression?). I should try one day the H-alpha with a colour camera to see what we get. But even there we have to adjust the RGB values in Firecapture, so it won't be objective H-alpha colour...
Certainly as Bob said, in all this trials we have to work on original files with no compression like .png or .tiff, if not we add one more factor of alteration of the 'objectivity' in colour.
It seems that this apparent loss of information or contrast in pure Red monochrome in comparison to Grayscale is due to our human perception of colours (the sensitivity of our eyes or the 'transcription' made by our brain).
I will try now the other method you proposed with Luminance or Colour layer to see what it gives as result.
Best wishes,
Alexandros
I tried this method (color Balance) myself as well before writing the post. The results differ depending on the sliders you use. Only midtones? All shadows, higlights and midtones? I don't know what is the 'correct' one. What did you use in your example?rigel123 wrote: ↑Mon Dec 12, 2022 8:45 pm
Yep, Robert gave you one of the more straight forward methods to get to a color you like. As you mentioned, you can also use an adjustment layer and play with levels on all three channels (RGB). It is a challenge to get that deep red color that you see visually and still see all of the nice details when working with an image. One thing I just thought of but haven't tried it yet is to make a deep red disk and then use a grayscale image as a layer to use as a Luminance layer like in deep sky processing.
I'll try posting this one more time, Chat has been really finicky today trying to get a post to actually post! Not sure what is up. I tried a couple of ways in PS to colorize a full disk I took on 12/4. This was the best method I found simply changing the mode to RGB and then used Color Balance and slid the Red slider all the way to the right and the G and B all the way to the left. Not too bad but I still see a loss of detail like I do anytime I colorize an image. Guess I'm a grayscale kind of guy!
Anyway, as you say, I also have the feeling that the Red we simulate in the computer looses quite some details in comparison to grayscale and also in comparison to other false colour (like the usual golden/yellow/orange many of us use for both H-alpha and white light. Even in Bob's cropped example, my eyes perceive less detail/contrast that in his grayscale (or is it just my personal impression?). I should try one day the H-alpha with a colour camera to see what we get. But even there we have to adjust the RGB values in Firecapture, so it won't be objective H-alpha colour...
Certainly as Bob said, in all this trials we have to work on original files with no compression like .png or .tiff, if not we add one more factor of alteration of the 'objectivity' in colour.
It seems that this apparent loss of information or contrast in pure Red monochrome in comparison to Grayscale is due to our human perception of colours (the sensitivity of our eyes or the 'transcription' made by our brain).
I will try now the other method you proposed with Luminance or Colour layer to see what it gives as result.
Best wishes,
Alexandros
Alexandros
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
Lunt 130MT+1800BF, C8 Ultima PEC+AstroSolar, Skywatcher Mount EQ6-R Pro
Baader Solar Prism, ZEISS Abbe Barlow 2x, Celestron Barlow 2x Ultima Series
ZWO ASI290MM, ZWO ASI1600MM Pro,
Greece
- rigel123
- Way More Fun to Share It!!
- Posts: 4648
- Joined: Thu Apr 08, 2021 9:26 pm
- Location: Mason, OH
- Has thanked: 5071 times
- Been thanked: 5575 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
I used the mid tone slider in my example. You could play with all three just to see what happens. I primarily stick with grayscale but every once in a while change things up with a bit of color.
Warren
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
- robert
- Way More Fun to Share It!!
- Posts: 2992
- Joined: Fri Oct 14, 2011 1:49 pm
- Location: N.W.Scotland
- Has thanked: 4 times
- Been thanked: 1042 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
I have done an adjustment using the .alv file that I use, on Bob's image (apologies Bob) to compare the three versions. I actually prefer the deep red image and the grey shows lots of detail more easily than that but the adjusted image in straw colour shows the detail with apparent increased contrast more easily and could be more beautiful than the plain grey to some people.
I take the point of jpg losses compared to png but disk space and upload speed are factors for me as well. I use .tif and .psd files for the processing stages before storing or posting images.
More to ponder!
Robert
I take the point of jpg losses compared to png but disk space and upload speed are factors for me as well. I use .tif and .psd files for the processing stages before storing or posting images.
More to ponder!
Robert
Last edited by robert on Wed Dec 14, 2022 2:21 pm, edited 1 time in total.
images and animations http://tinyurl.com/h5bgoso
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
- rigel123
- Way More Fun to Share It!!
- Posts: 4648
- Joined: Thu Apr 08, 2021 9:26 pm
- Location: Mason, OH
- Has thanked: 5071 times
- Been thanked: 5575 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
When I did my adjustments I worked on the TIF version, but of course converted it into a jpg for posting.
Warren
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
Lunt LS60T DS
Orion ED80T CF
Lunt CaK BF1200
Lunt WL Wedge
ASI174MM
Skyris 236M
https://www.astrobin.com/users/rigel123/
- robert
- Way More Fun to Share It!!
- Posts: 2992
- Joined: Fri Oct 14, 2011 1:49 pm
- Location: N.W.Scotland
- Has thanked: 4 times
- Been thanked: 1042 times
- Contact:
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Alexandros
I have tried to extract what the .alv file that I use actually does, just in case it is of use. It opens the R, G and B channels in "image adjust" and alters the value in the middle box for R to 1.2 approx, G to 0.5 approx and B to 0.1.
Robert
I have tried to extract what the .alv file that I use actually does, just in case it is of use. It opens the R, G and B channels in "image adjust" and alters the value in the middle box for R to 1.2 approx, G to 0.5 approx and B to 0.1.
Robert
images and animations http://tinyurl.com/h5bgoso
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
2023 images https://www.flickr.com/photos/69734017@ ... 0304905278
2022 images https://www.flickr.com/photos/69734017@ ... 0295810277
2021 images https://tinyurl.com/d7jbz3sw
ED80. ED100. SW200. Celestron-150mm-PST mod. C9.25. ES127
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265
-
- The Sun?
- Posts: 1
- Joined: Thu Apr 20, 2023 7:24 am
- Been thanked: 1 time
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Yes, it is possible to false colorize a grayscale H-alpha photo objectively in RGB. False colorization involves mapping grayscale values to colors, which can be done using a variety of algorithms.
One common approach is to develop a color table, where each grayscale value is mapped to a corresponding color. This approach can be subjective if the color table is chosen based on personal preferences or artistic considerations. An objective approach can be achieved by using a scientifically-based color map, such as the viridis or magma color maps, which are designed to perceptually optimize color use to represent data. These color maps have been shown to be effective in displaying scientific data and are widely used in the scientific community.
Therefore, by using a scientifically-based color map, it is possible to objectively falsely colorize a grayscale H-alpha photo in RGB.
One common approach is to develop a color table, where each grayscale value is mapped to a corresponding color. This approach can be subjective if the color table is chosen based on personal preferences or artistic considerations. An objective approach can be achieved by using a scientifically-based color map, such as the viridis or magma color maps, which are designed to perceptually optimize color use to represent data. These color maps have been shown to be effective in displaying scientific data and are widely used in the scientific community.
Therefore, by using a scientifically-based color map, it is possible to objectively falsely colorize a grayscale H-alpha photo in RGB.
[https://agnisolar.com/][/solar power system]
- Montana
- Librarian
- Posts: 32022
- Joined: Mon Oct 17, 2011 5:25 pm
- Location: Cheshire, UK
- Has thanked: 13555 times
- Been thanked: 6963 times
Re: Is it possible to false colourize a grayscale H-alpha photo objectively in RGB?
Hi Robert,
I have been trying to follow your instructions above to colour red in Photoshop. I change the mode to RGB, then go to the layers tab (bottom right) then click on the channels tab and I see 4 channels with eyes on RGB, Red, Green, Blue. However, how do I click on them and 'fill them with black' ?
Many thanks for your help
Alexandra
I have been trying to follow your instructions above to colour red in Photoshop. I change the mode to RGB, then go to the layers tab (bottom right) then click on the channels tab and I see 4 channels with eyes on RGB, Red, Green, Blue. However, how do I click on them and 'fill them with black' ?
Many thanks for your help
Alexandra