Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Using colour camera for the RGB and mono camera for luminance?

CMOS Reflector
  • Please log in to reply
26 replies to this topic

#1 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 28 April 2021 - 12:14 PM

I have a Skywatcher 200PDS 8" Newtonian and ZWO ASI 2600MC colour camera which I am pleased with. I have noticed that mono camera users observe for far longer with the luminance channel than for RGB as the eye sees more details in the black and white compared with colour.

So would there be much advantage of using a mono camera in luminance to enhance the detail picked up by the colour camera, as it is more sensitive and has higher resolution for the same camera type than the colour camera?

If that worked, could I use a cheaper smaller chip size camera, if the stacking software could resize the images to fit together? Deep Sky Stacker which is what I use couldn't  as I understand it.



#2 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,159
  • Joined: 27 Oct 2014

Posted 28 April 2021 - 12:39 PM

I have a Skywatcher 200PDS 8" Newtonian and ZWO ASI 2600MC colour camera which I am pleased with. I have noticed that mono camera users observe for far longer with the luminance channel than for RGB as the eye sees more details in the black and white compared with colour.

So would there be much advantage of using a mono camera in luminance to enhance the detail picked up by the colour camera, as it is more sensitive and has higher resolution for the same camera type than the colour camera?

If that worked, could I use a cheaper smaller chip size camera, if the stacking software could resize the images to fit together? Deep Sky Stacker which is what I use couldn't  as I understand it.

Perfectly good idea.  But a smaller chip would throw away some of the field of view from your larger one.  I'd, instead, get a matched set.

 

Which I did with my 2 183s.  Which are in the process of being replaced by 2600s.


  • psandelle and Hobby Astronomer like this

#3 adosaj

adosaj

    Explorer 1

  • *****
  • Posts: 75
  • Joined: 04 Jul 2017

Posted 28 April 2021 - 01:24 PM

Why build one when you can have two at twice the price! - S.R. Hadden

grin.gif


  • John Miele likes this

#4 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 29 April 2021 - 03:10 AM

I did enjoy that film. Actually is more than twice the price as the mono version is quite a bit more expensive despite not having the bayer filter. Apart from being an expensive option, has anyone tried using a mono and colour camera of the same spec to get RGB and luminance images? 


Edited by Moorefam, 29 April 2021 - 03:12 AM.


#5 8472

8472

    Vostok 1

  • -----
  • Posts: 151
  • Joined: 20 Jan 2018

Posted 29 April 2021 - 03:36 AM

Yes, that is exactly what I do.

RGB with my Olympus M4/3 mirrorless.

Mono/luminance with my QHY294M.

Although they are both M4/3 chips, the QHY has a slightly wider FOV and either smaller or larger pixels, depending on which mode it operates.

Not enough to cause problems, though.

You are correct in that DSS will not register dissimilar image scales, but there is alternative AP software out there that will.
  • psandelle, bobzeq25 and Pelle like this

#6 TareqPhoto

TareqPhoto

    Fly Me to the Moon

  • -----
  • Posts: 6,800
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 29 April 2021 - 11:37 AM

I am planning to do that, the only missing part in this idea for me is the scopes.



#7 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 29 April 2021 - 11:38 AM

That's interesting and I would be interested in what improvement in image quality you found compared with using your Olympus alone.



#8 8472

8472

    Vostok 1

  • -----
  • Posts: 151
  • Joined: 20 Jan 2018

Posted 30 April 2021 - 04:37 AM

The improvement is night and day, no doubt about it.

The beauty of it is that, if you have any, you can revisit your old OSC data from way back, (provided you didn't destroy colour balance by using light pollution filters) and transform them with luminance.

You will be be surprised how poor your RGB can be (within reason) when you augment it with good mono data.

#9 the Elf

the Elf

    Soyuz

  • *****
  • Posts: 3,509
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 30 April 2021 - 03:32 PM

I did color with a stock Canon T3i and added L from a mono modded T3i. If you have a mono and a color version of the same chip it is a great idea. Now that I have the superior T7i there is no point in adding L data from an old and inferior chip.

Here is an example:

https://www.elf-of-l...j_M13_2019.html

Despite the relatively low over all integration time there is a lot of detail in the image considering it is taken with DSLRs. Go for it! You will be surprised.



#10 TareqPhoto

TareqPhoto

    Fly Me to the Moon

  • -----
  • Posts: 6,800
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 30 April 2021 - 04:30 PM

I did color with a stock Canon T3i and added L from a mono modded T3i. If you have a mono and a color version of the same chip it is a great idea. Now that I have the superior T7i there is no point in adding L data from an old and inferior chip.

Here is an example:

https://www.elf-of-l...j_M13_2019.html

Despite the relatively low over all integration time there is a lot of detail in the image considering it is taken with DSLRs. Go for it! You will be surprised.

In the description it mentioned Canon T3i and not T7i.

 

Good comment, which means i should do a lot with my OSC i ordered and very soon to receive it, it is a new generation camera and cooled, based on Sony IMX571 sensor which is found in cameras such as ASI2600 or QHY268, can't wait to see, and i do have mono cameras, i want to add L and Ha from mono to that camera data.


  • skyguy55 likes this

#11 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 01 May 2021 - 10:00 AM

Yes that's a lot of detail in M13 so well done indeed.

Thanks for all the help.

What I am not sure of is how I would include the Luminance in the stacking? I suppose I would have to do it in Photoshop? 

I am also not certain what would happen when stacking a luminance channel on top of a colour RGB image as say the fainter and fainter blue outer rim of a galaxy

disappears in the noise but the luminance channel continues to reveal detail. I suppose the blue colour would gradually merge into mono shades but with better detail all over the image?



#12 the Elf

the Elf

    Soyuz

  • *****
  • Posts: 3,509
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 01 May 2021 - 12:44 PM

The processing is done the same way as LRGB with a mono. Each set is stacked independently with it's own set of flats, darks and bias. L and color is processed independently. Usually you would star align all (both) sets to the same sub, probably a good light frame. For example in PI it is not a problem to align stacks to each other.

 

In post L is being sharpened for maximum detail while color can be denoised a lot. For L you must not stretch too much. For a mono image you would go for a lot of contrast, for the combination you should stay away from too bright areas because the result will lack color. Which software do you use for post processing?



#13 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 01 May 2021 - 05:19 PM

Thanks for the information. Apart from DSS I use Photoshop mainly and Lightroom a little. 



#14 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,159
  • Joined: 27 Oct 2014

Posted 01 May 2021 - 06:58 PM

 Usually you would star align all (both) sets to the same sub, probably a good light frame. For example in PI it is not a problem to align stacks to each other.

 

 

What I've seen says there's a minor noise advantage to aligning all subs to a single reference, as compared to aligning stacks (which, as you say, is a possibility in PI).

 

I see no good reason to not just align subs to a common reference frame. so that's what I do.
 


  • jdupton likes this

#15 the Elf

the Elf

    Soyuz

  • *****
  • Posts: 3,509
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 02 May 2021 - 06:24 AM

Each sub is resampled for star alignment applying a rotation, translation and sometimes a small change of size due to focus shift. The resample is a Lanczos filter that is somewhat edge preserving to edge enhancing depending on the parameters used. If you align stacks one of them is resampled for a second time. If you rotate the image by an odd angle another resample occurs. I tend to keep the amount of these as low as possible.



#16 JamesTX

JamesTX

    Messenger

  • *****
  • Posts: 424
  • Joined: 27 Aug 2017
  • Loc: Texas

Posted 02 May 2021 - 04:29 PM



I have a Skywatcher 200PDS 8" Newtonian and ZWO ASI 2600MC colour camera which I am pleased with. I have noticed that mono camera users observe for far longer with the luminance channel than for RGB as the eye sees more details in the black and white compared with colour.

So would there be much advantage of using a mono camera in luminance to enhance the detail picked up by the colour camera, as it is more sensitive and has higher resolution for the same camera type than the colour camera?

If that worked, could I use a cheaper smaller chip size camera, if the stacking software could resize the images to fit together? Deep Sky Stacker which is what I use couldn't  as I understand it.


My latest image of M101 taken with a 294mm I needed just 1 hour of RGB each and took 10.5 hours on Lum:

 

get.jpg?insecure

 

My previous image of the NGC 3628 has 14 hours with a OSC (asi533) and 11 hours of lum from the 294mm.  I shot it this way because I received the 294mm after I already started gathering data with the asi533:

 

get.jpg?insecure

 

 

 

Looking at these two shots.. If I had the 294mm when starting the Hamburger.. and shot just 1 hour each of RGB and spent the remaining time used with the OSC on Lum with the 294 instead.. I'd be over 20 hours or lum... or maybe I wouldn't even need that much time on it.  I guess what I'm trying to say is that the amount of color data needed to work with lum is so low.. i don't think it really makes much sense using a separate OSC and lum camera. 

 

I use to think if I had a dual setup.. run one with OSC and the other with Lum that would be an awesome and efficient way to get lots of data on a single target.. but so far my experiences are telling me I'd be much better off getting significantly more lum data than close to equal amount of time with a mono/lum and OSC.

 

So with all that said.. if you already have a OSC I would stick with that.  If you are thinking about going Mono.. might as well go pure mono.  Maybe as a intermediate step.. assuming mono is an eventuality you can grab the mono with a lum and Ha filter.. and get the color data with the OSC until you have accumulated all the filters.. that would be a viable path.

 

Hopefully this makes sense.. otherwise Ive just muddied the waters further smile.gif


  • hornjs likes this

#17 TareqPhoto

TareqPhoto

    Fly Me to the Moon

  • -----
  • Posts: 6,800
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 02 May 2021 - 07:09 PM

My latest image of M101 taken with a 294mm I needed just 1 hour of RGB each and took 10.5 hours on Lum:

 

get.jpg?insecure

 

My previous image of the NGC 3628 has 14 hours with a OSC (asi533) and 11 hours of lum from the 294mm.  I shot it this way because I received the 294mm after I already started gathering data with the asi533:

 

get.jpg?insecure

 

 

 

Looking at these two shots.. If I had the 294mm when starting the Hamburger.. and shot just 1 hour each of RGB and spent the remaining time used with the OSC on Lum with the 294 instead.. I'd be over 20 hours or lum... or maybe I wouldn't even need that much time on it.  I guess what I'm trying to say is that the amount of color data needed to work with lum is so low.. i don't think it really makes much sense using a separate OSC and lum camera. 

 

I use to think if I had a dual setup.. run one with OSC and the other with Lum that would be an awesome and efficient way to get lots of data on a single target.. but so far my experiences are telling me I'd be much better off getting significantly more lum data than close to equal amount of time with a mono/lum and OSC.

 

So with all that said.. if you already have a OSC I would stick with that.  If you are thinking about going Mono.. might as well go pure mono.  Maybe as a intermediate step.. assuming mono is an eventuality you can grab the mono with a lum and Ha filter.. and get the color data with the OSC until you have accumulated all the filters.. that would be a viable path.

 

Hopefully this makes sense.. otherwise Ive just muddied the waters further smile.gif

For both images you used C8 with 0.7x reducer, that is slow to my dictionary, i consider F5 to be maximum speed, you didn't calculate how long we will have with OSC on scopes that are fast between say F2 up to F5, i can use my OSC on my 6" Newt which is F4, and if i can use something like 0.9x that will be F3.6, i saw 0.73x reducer corrector so that will crank speed to F2.9, imagine how much data i can get with that for RGB filters or OSC anyway.

 

It is not about a competition between RGB filters and OSC, if i can do say 5 hours with OSC that i can do for 1 hour total of RGB i still go with OSC, why? Because from all time i spent last year i faced situations where i always miss one filter or two out of three, so it will lose all data then, while with OSC and i fix the temp and focus i just go and sleep and forget, for me 3-6 hours good data with OSC is better than 100% quality red filter data and 50% green and 10% blue, pointless, that is a waste really.

 

And if we spend hours under stars, why we have to make OSC any faster than LRGB? I wanted OSC for times when it is overkill for RGB anyway, sometimes OSC can do enough, i saw so so many results from OSC at all sensor sizes and they are amazing or mind blowing, so technology is improved now and it can be fine, and we already adding Lum or Ha from mono which holds the details, that is something to consider, talking about myself i wasn't planning to replace the mono completely, it was mainly to add RGB from one camera than from filters making sure each filter is perfect, if i didn't see results from OSC then i won't think about it.



#18 JamesTX

JamesTX

    Messenger

  • *****
  • Posts: 424
  • Joined: 27 Aug 2017
  • Loc: Texas

Posted 03 May 2021 - 05:54 PM

For both images you used C8 with 0.7x reducer, that is slow to my dictionary, i consider F5 to be maximum speed, you didn't calculate how long we will have with OSC on scopes that are fast between say F2 up to F5, i can use my OSC on my 6" Newt which is F4, and if i can use something like 0.9x that will be F3.6, i saw 0.73x reducer corrector so that will crank speed to F2.9, imagine how much data i can get with that for RGB filters or OSC anyway.

 

It is not about a competition between RGB filters and OSC, if i can do say 5 hours with OSC that i can do for 1 hour total of RGB i still go with OSC, why? Because from all time i spent last year i faced situations where i always miss one filter or two out of three, so it will lose all data then, while with OSC and i fix the temp and focus i just go and sleep and forget, for me 3-6 hours good data with OSC is better than 100% quality red filter data and 50% green and 10% blue, pointless, that is a waste really.

 

And if we spend hours under stars, why we have to make OSC any faster than LRGB? I wanted OSC for times when it is overkill for RGB anyway, sometimes OSC can do enough, i saw so so many results from OSC at all sensor sizes and they are amazing or mind blowing, so technology is improved now and it can be fine, and we already adding Lum or Ha from mono which holds the details, that is something to consider, talking about myself i wasn't planning to replace the mono completely, it was mainly to add RGB from one camera than from filters making sure each filter is perfect, if i didn't see results from OSC then i won't think about it.

Whatever works for you.  I was just sharing my opinion.  Good luck with your new camera/strategy  :)



#19 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 04 May 2021 - 11:07 AM

Thanks again for the helpful replies.

Janes TX I really like your NGC3638 and M101 images, they are impressive and sharp. Regarding M101 I like the prominent red Ha cluster regions in the spirals but in my M101 image even after 14.5 hrs subs the red Ha regions are not striking. I have noticed this with other galaxy comparisons.

Apart from more detailed images I am particularly interested in bringing out out these Ha regions that are proving difficult to see clearly at present. Would an Ha filter do the job?

Also my ZWO ASI 2600MC has an IR block but no UV block. Should I have a UV block filter permanently in place as without it I understand I can have bloated stars as the UV light focusses in a different plane to the visible? My stars are not as small as I would like despite careful focussing and a main mirror fan which I recently fitted though it seems to make FHMW figures greater, expelling air, maybe less so blowing up the tube.

I only started astrophotography a year ago so these are early days.. 

I have an example of my M101 image on my webpage at www.moorefam.plus.com. This will show what I mean about the Ha regions not being prominent. 



#20 JamesTX

JamesTX

    Messenger

  • *****
  • Posts: 424
  • Joined: 27 Aug 2017
  • Loc: Texas

Posted 04 May 2021 - 11:37 AM

Thanks again for the helpful replies.

Janes TX I really like your NGC3638 and M101 images, they are impressive and sharp. Regarding M101 I like the prominent red Ha cluster regions in the spirals but in my M101 image even after 14.5 hrs subs the red Ha regions are not striking. I have noticed this with other galaxy comparisons.

Apart from more detailed images I am particularly interested in bringing out out these Ha regions that are proving difficult to see clearly at present. Would an Ha filter do the job?

Also my ZWO ASI 2600MC has an IR block but no UV block. Should I have a UV block filter permanently in place as without it I understand I can have bloated stars as the UV light focusses in a different plane to the visible? My stars are not as small as I would like despite careful focussing and a main mirror fan which I recently fitted though it seems to make FHMW figures greater, expelling air, maybe less so blowing up the tube.

I only started astrophotography a year ago so these are early days.. 

I have an example of my M101 image on my webpage at www.moorefam.plus.com. This will show what I mean about the Ha regions not being prominent. 

Thanks David!

 

In regards to the asi2600mc, yes you want to block both UV and IR.  I don't have a 2600mc myself but I was under the impression that it blocked both with its cover.  The asi533 which I have doesn't block either and I use a luminance filter for that purpose. 

 

For the Ha regions, yes you need an Ha filter.  Actually, if you have something like the L-Extreme or triad you can use that instead of an Ha filter.

 

The procedure is pretty straight forward.  I use Pixinsight so I can explain the basic steps in PI.

 

- Assuming you use something like the L-Extreme for Ha, split the L-extreme data into separate RGB channels.  Label the red as Ha and discard the green and blue.

- With your broadband data.. also split it out into separate RGB channels.

- Then use pixel math to combine the Ha and Red channel.  Lately Ive been using 40% Ha and 60% Red.

- Then recombine your RGB data with the lrgb combination tool.. using the combined Red+Ha image in the red channel

 

Sometimes the Ha is too strong.  You can use the Ha frame to create a mask to isolate the Ha regions when you continue your post processing.

 

I used the above method with this shot of M33, taken with my 70mm refractor (sv70t), the asi533 and the L-Extreme for the Ha data:

 

get.jpg?insecure

 

I really should post a video on how I do this.  On my youtube channel I posted a video comparing my previous attempts at M101 with this latest attempt.  It's fun to see the progression you make after a couple years of experience. 

 

btw.. your M51 shot looks fantastic! 


Edited by JamesTX, 04 May 2021 - 11:45 AM.

  • bobzeq25 likes this

#21 Moorefam

Moorefam

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 12 Jun 2020

Posted 04 May 2021 - 12:56 PM

James,

I had already watched your M101 journey Youtube video and noticed how your latest image was the sharpest. It's an interesting video and informative do thanks for sharing it.

I don't have PI but with what you have told me and by also looking at an Astro Backyard video I expect I shall be able to manage eventually. 2" Ha filters are rather expensive though but it is an expensive hobby.

I have managed to answer my own question about the ZWO ASI 2600MC. It has v sharp cut off at 400nM and 700nM and 400nM is the limit for the UV wavelengths so I don't need to worry about star bloat too much though I have plenty on a image I took of the Pleaides.  I had the gain at 240 with 2 min subs and have since lowered it to 100. 

Thanks for the comment about M51. Perhaps I should give it the light of day and put it on Astrobin? So far my images have only been seen by family and friends and my camera club so a comment from someone experienced like yourself is really encouraging. 


Edited by Moorefam, 04 May 2021 - 12:59 PM.

  • JamesTX likes this

#22 APshooter

APshooter

    Skylab

  • *****
  • Posts: 4,349
  • Joined: 14 Oct 2013
  • Loc: Camby, In.

Posted 04 May 2021 - 03:56 PM

Kind of late to the party but I've been using separate OSC and L cameras for years. I'm very pleased with the results so far, especiallyon my M31 and M33 shots. I need to get the 2600mm to go along with my 2600MC to really take advantage of the process.

#23 TareqPhoto

TareqPhoto

    Fly Me to the Moon

  • -----
  • Posts: 6,800
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 04 May 2021 - 11:05 PM

Kind of late to the party but I've been using separate OSC and L cameras for years. I'm very pleased with the results so far, especiallyon my M31 and M33 shots. I need to get the 2600mm to go along with my 2600MC to really take advantage of the process.

Show me something about that, please



#24 CloudGazer

CloudGazer

    Lift Off

  • -----
  • Posts: 9
  • Joined: 22 Oct 2020

Posted 05 May 2021 - 08:35 AM

- With your broadband data.. also split it out into separate RGB channels.

- Then use pixel math to combine the Ha and Red channel.  Lately Ive been using 40% Ha and 60% Red.

Another way in Pixelmath is to use max(OSC, Ha) - Sometimes it's necessary with a factor on one of them to get the right balance.  This typically leaves star colors intact, but adds Ha nebulosity nicely.

 

Though first you need to put the Ha data in the Red channel of an empty OSC image - Use Pixelmath for that too.

 

I learned that from the Elf's YouTube videos, by the way!


  • JamesTX likes this

#25 gnnyman

gnnyman

    Lift Off

  • -----
  • Posts: 5
  • Joined: 21 Aug 2019
  • Loc: Lower Austria

Posted 05 May 2021 - 08:55 AM

I am planning to mix L from ASI183MM Pro with RGB from ASI183MC Pro - this should be OK, I guess. However, I found out, that if I use the wieghted batch preprocessing in PI and want to use Dark frames from the one together with Flatframes from the other plus some RGB with some L data, it won´t work. Have you got some similar experiences? I did (my fault) not pay attention when I started the batch process - I just took Darks, Flats, Bias and Lights which matched in Gain and Offset (and where necessary also in time), but did not look at them - some were RGB, some 16bit Mono....
It means, I think, that I have to process the RGB data completely separate and then separate also the 16bit mono data and later mix them in the script LNRGB...I think, that should work.

Any ideas and suggestions?
Thanks

Georg

 

Equipment: Lacerta FotoNewton 254/1250Q plus ASI183MC Pro and/or ASI183MM Pro with manual filter holder 2" for Ha, OIII etc filters, SW: NINA Nightly




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics





Also tagged with one or more of these keywords: CMOS, Reflector



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics