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

LRGB vs L-RGB

  • Please log in to reply
11 replies to this topic

#1 moxican

moxican

    Viking 1

  • *****
  • topic starter
  • Posts: 704
  • Joined: 13 Nov 2011
  • Loc: Toronto, Canada

Posted 24 January 2020 - 07:08 PM

I have been searching online to get some answer but info on the subject is limited. LRGB imaging is straight forward, by L-RGB I mean shooting luminosity data with a mono camera and getting RGB would come from a OSC camera.

Let me put down a detailed scenario:

 

Set-up:

 

Luminosity: 

Refractor at 500mm focal length

ASI178mm Cool (2.4 micron)

 

RGB:

Camera lens at 500mm focal length

Nikon D5100 (4.8 micron)

 

 

For sake of simplicity let's assume that guiding and dithering is at sufficient level, and disregard the difference in field of view.

 

I am theorizing that I could use the RGB data for color and Luminance from the ZWO for luminosity. The pixel size is of the DSLR is twice as large as the ZWO, 4.8 vs 2.4 respectively. If I stack the OSC data with 2x drizzle in DSS that would give me the same sampling resolution (0.99"/pixel) as the ASI178. This way I could crop out the "area of interest" from the OSC data, align with luminosity and combine later down process flow. This way it would save me some time on the field, not needing to image separe R, G, and B data. 

I've read here on CN somewhere I think ( by user ccs_hello if I remember right) that the quality of the separate color data also makes a difference, although many imagers tend to shoot their RGB binning it 2x2 in order to save some time on the field. Consequently, the same theory would fly if I bin the luminosity data 2x2 and combine down the processing flow.

Would this set-up work?

What I am really after is saving imaging time. Up here north in this season I get 1-2 clear nights a month at best. I want to make the best of those rare nights.


Edited by moxican, 24 January 2020 - 07:12 PM.


#2 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,558
  • Joined: 27 Oct 2014

Posted 24 January 2020 - 08:09 PM

I have been searching online to get some answer but info on the subject is limited. LRGB imaging is straight forward, by L-RGB I mean shooting luminosity data with a mono camera and getting RGB would come from a OSC camera.

Let me put down a detailed scenario:

 

Set-up:

 

Luminosity: 

Refractor at 500mm focal length

ASI178mm Cool (2.4 micron)

 

RGB:

Camera lens at 500mm focal length

Nikon D5100 (4.8 micron)

 

 

For sake of simplicity let's assume that guiding and dithering is at sufficient level, and disregard the difference in field of view.

 

I am theorizing that I could use the RGB data for color and Luminance from the ZWO for luminosity. The pixel size is of the DSLR is twice as large as the ZWO, 4.8 vs 2.4 respectively. If I stack the OSC data with 2x drizzle in DSS that would give me the same sampling resolution (0.99"/pixel) as the ASI178. This way I could crop out the "area of interest" from the OSC data, align with luminosity and combine later down process flow. This way it would save me some time on the field, not needing to image separe R, G, and B data. 

I've read here on CN somewhere I think ( by user ccs_hello if I remember right) that the quality of the separate color data also makes a difference, although many imagers tend to shoot their RGB binning it 2x2 in order to save some time on the field. Consequently, the same theory would fly if I bin the luminosity data 2x2 and combine down the processing flow.

Would this set-up work?

What I am really after is saving imaging time. Up here north in this season I get 1-2 clear nights a month at best. I want to make the best of those rare nights.

The setup could work just fine.  But what it won't do is get you data faster.  The 2X2 Baader matrix filter, using dyed glass, was designed for terrestrial use in inexpensive cameras.  It's less efficient than full sized R,G,B interference filters.

 

If you do go that way, I'd just shoot luminance binned 1X1.  The larger numerical imaging scale of the RGB would resemble binning the color data 2X2.  Your eyes see detail in luminance.  LRGB imaging (either way) is similar, in that what you're basically doing is painting the base luminance image with RGB.  It's a clever trick, based on how your eyes see.

 

I'm planning on doing L-RGB (using a one shot color astro camera) myself.  I have a C8 RASA, which does not allow for the use of a filter wheel.  There's nothing fundamentally wrong with the idea.


Edited by bobzeq25, 24 January 2020 - 08:18 PM.


#3 moxican

moxican

    Viking 1

  • *****
  • topic starter
  • Posts: 704
  • Joined: 13 Nov 2011
  • Loc: Toronto, Canada

Posted 24 January 2020 - 08:50 PM

The setup could work just fine.  But what it won't do is get you data faster.  The 2X2 Baader matrix filter, using dyed glass, was designed for terrestrial use in inexpensive cameras.  It's less efficient than full sized R,G,B interference filters.

 

If you do go that way, I'd just shoot luminance binned 1X1.  The larger numerical imaging scale of the RGB would resemble binning the color data 2X2.  Your eyes see detail in luminance.  LRGB imaging (either way) is similar, in that what you're basically doing is painting the base luminance image with RGB.  It's a clever trick, based on how your eyes see.

 

I'm planning on doing L-RGB (using a one shot color astro camera) myself.  I have a C8 RASA, which does not allow for the use of a filter wheel.  There's nothing fundamentally wrong with the idea.

I am planning to shoot the two set-up side by side. That is what I mean by saving time. So, in those terms I'd save time due to exposing for OSC the same time as shooting for Luminosity.



#4 whwang

whwang

    Gemini

  • *****
  • Posts: 3,019
  • Joined: 20 Mar 2013

Posted 25 January 2020 - 12:39 AM

It should work.  You are not the first one doing something like this.  

 

Don't bin the data during readout.  Image at the native resolutions of both cameras, and let the post-processing software handle the difference in pixel size.


  • bobzeq25 likes this

#5 moxican

moxican

    Viking 1

  • *****
  • topic starter
  • Posts: 704
  • Joined: 13 Nov 2011
  • Loc: Toronto, Canada

Posted 25 January 2020 - 12:49 PM

Will the sharpness of the OSC camera data matter much?

There are some lenses that are softer than others. If details will be defined by luminosity only, then RGB should only bring color. Is that correct?



#6 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,558
  • Joined: 27 Oct 2014

Posted 25 January 2020 - 02:34 PM

Will the sharpness of the OSC camera data matter much?

There are some lenses that are softer than others. If details will be defined by luminosity only, then RGB should only bring color. Is that correct?

It will matter some.  Not as much as some might think.

 

As usual in DSO AP, this is quite complicated.  Theoretical musings can only get you so far.  Sometimes you just have to try stuff.  <smile>



#7 calypsob

calypsob

    Aurora

  • *****
  • Posts: 4,978
  • Joined: 20 Apr 2013

Posted 25 January 2020 - 03:35 PM

I have been searching online to get some answer but info on the subject is limited. LRGB imaging is straight forward, by L-RGB I mean shooting luminosity data with a mono camera and getting RGB would come from a OSC camera.

Let me put down a detailed scenario:

 

Set-up:

 

Luminosity: 

Refractor at 500mm focal length

ASI178mm Cool (2.4 micron)

 

RGB:

Camera lens at 500mm focal length

Nikon D5100 (4.8 micron)

 

 

For sake of simplicity let's assume that guiding and dithering is at sufficient level, and disregard the difference in field of view.

 

I am theorizing that I could use the RGB data for color and Luminance from the ZWO for luminosity. The pixel size is of the DSLR is twice as large as the ZWO, 4.8 vs 2.4 respectively. If I stack the OSC data with 2x drizzle in DSS that would give me the same sampling resolution (0.99"/pixel) as the ASI178. This way I could crop out the "area of interest" from the OSC data, align with luminosity and combine later down process flow. This way it would save me some time on the field, not needing to image separe R, G, and B data. 

I've read here on CN somewhere I think ( by user ccs_hello if I remember right) that the quality of the separate color data also makes a difference, although many imagers tend to shoot their RGB binning it 2x2 in order to save some time on the field. Consequently, the same theory would fly if I bin the luminosity data 2x2 and combine down the processing flow.

Would this set-up work?

What I am really after is saving imaging time. Up here north in this season I get 1-2 clear nights a month at best. I want to make the best of those rare nights.

If you can possibly obtain a second 178 is osc, I think the results would be superior to dealing with the dslr data.  

Alternatively you could get another d5100, they are pretty cheap used.  You could then just integrate data from both cameras.

Integrating dual dslr's gives you a fierce ability to reject outliers, especially if they both sample at the same pixels scale. In my experience there are no residual hot pixels and the sample size is so dense, you have almost no leftover noise after 3-4 hours. I have never tried dual cooled cmos but the results would obviously be even more superior due to the regulated temperatures. 


Edited by calypsob, 25 January 2020 - 03:36 PM.


#8 moxican

moxican

    Viking 1

  • *****
  • topic starter
  • Posts: 704
  • Joined: 13 Nov 2011
  • Loc: Toronto, Canada

Posted 25 January 2020 - 10:51 PM

If you can possibly obtain a second 178 is osc, I think the results would be superior to dealing with the dslr data.  

Alternatively you could get another d5100, they are pretty cheap used.  You could then just integrate data from both cameras.

Integrating dual dslr's gives you a fierce ability to reject outliers, especially if they both sample at the same pixels scale. In my experience there are no residual hot pixels and the sample size is so dense, you have almost no leftover noise after 3-4 hours. I have never tried dual cooled cmos but the results would obviously be even more superior due to the regulated temperatures. 

I understand what you are saying. Getting another D5100 would not get me closer towards what I want. I'd like to get high resolution images and with the scope (AT90 EDT) of 600mm native focal length and 2.4 micron pixel size of he ASI178mm (or ASI183mm) would give me twice the sampling that I get with my D5100.

Getting a second camera as the ASI178mc makes a bit more sense.

What do you have against using color data from the D5100. If I drizzle 2x it would give me the same pixel sampling as the ASI178mm.



#9 whwang

whwang

    Gemini

  • *****
  • Posts: 3,019
  • Joined: 20 Mar 2013

Posted 26 January 2020 - 12:58 AM

Will the sharpness of the OSC camera data matter much?
There are some lenses that are softer than others. If details will be defined by luminosity only, then RGB should only bring color. Is that correct?


That’s roughly right. You still don’t want the two to have very different resolutions though.

#10 moxican

moxican

    Viking 1

  • *****
  • topic starter
  • Posts: 704
  • Joined: 13 Nov 2011
  • Loc: Toronto, Canada

Posted 31 January 2020 - 08:48 PM

I am also concerned about interpolation for the DSLR camera. Interpolation will make the OSC image blurry and the stars will look more bloated than they'll look on the mono luminosity image. Won't that cause a problem for the stars?



#11 whwang

whwang

    Gemini

  • *****
  • Posts: 3,019
  • Joined: 20 Mar 2013

Posted 01 February 2020 - 03:34 AM

I don't think it will hurt that much.

 

One thing you may do is to search for pictures of the same object on the internet, taken at lower resolution and higher resolution.  You turn the high-res one into mono, and do an LRGB merge with the low-res color one, to see what happens.  As long as you can find internet images with similar characteristics to what you are going to get by yourself, this test should give you a good idea about what you will see.



#12 xiga

xiga

    Viking 1

  • *****
  • Posts: 531
  • Joined: 08 Oct 2012
  • Loc: Northern Ireland

Posted 12 February 2020 - 06:41 AM

I am also concerned about interpolation for the DSLR camera. Interpolation will make the OSC image blurry and the stars will look more bloated than they'll look on the mono luminosity image. Won't that cause a problem for the stars?


As long as your subs are well-dithered, and you get a good sized number of them, then you can always do a Bayer Drizzle integration and avoid interpolation altogether.


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






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics