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

Comprehensive LRGB vs. OSC discussion?

Astrophotography CMOS DSO Filters
  • Please log in to reply
5 replies to this topic

#1 Eyeroll1952

Eyeroll1952

    Sputnik

  • *****
  • topic starter
  • Posts: 40
  • Joined: 05 Mar 2021
  • Loc: Berthoud, CO

Posted 22 June 2022 - 08:17 AM

Is anyone aware of a comprehensive discussion of the relative merits of LRGB vs. OSC imaging? I am aware of multiple interesting CN threads and such but feel they are not necessarily comprehensive as they typically to a large extent discuss the issue in photon counting terms (or maybe better called counting photon terms). Or they bring up issues of chance: if you are unlucky with weather you could lose e.g. all your red subs. But the counter argument to that is to collect more data on a different day. I am particularly interested in two questions:

 

1. Given that blue and red dominates green for many DSO targets, does that suggest one should optimally spend less time collecting on the green channel, or perhaps one should collect more, or perhaps it doesn’t matter in practice?

2. When looking at spectral response curves for OSC sensors (see typical picture below left) it is striking how much bleed there is between color channels compared with RGB filters that have much reduced bleed (right). One example implication is that a non-trivial amount of Ha should show up in the green and even blue channels. The question then is whether this channel “leakage” makes a substantial difference in typical images.

 

Yes, I know in the end image processing and colorization are to a significant extent a matter of aesthetics but am curious to understand if people have poked at issues like this?

 

Cheers,

 

Paul

 

OSC spectrum.jpg LRGB spectra.jpg

 



#2 jdupton

jdupton

    Skylab

  • *****
  • Posts: 4,443
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 22 June 2022 - 08:43 AM

Paul,

 

1. Given that blue and red dominates green for many DSO targets, does that suggest one should optimally spend less time collecting on the green channel, or perhaps one should collect more, or perhaps it doesn’t matter in practice?

2. When looking at spectral response curves for OSC sensors (see typical picture below left) it is striking how much bleed there is between color channels compared with RGB filters that have much reduced bleed (right). One example implication is that a non-trivial amount of Ha should show up in the green and even blue channels. The question then is whether this channel “leakage” makes a substantial difference in typical images.

 

   If you have already read the other main CloudyNights Mono to OSC comparison threads, maybe your specific questions will be answered by looking at the responses from two specific CN members on those subjects. Frank (freestar8n), the author of MetaGuide and Mark (sharkmelley) have written a lot about both the importance of the Green channel in astro-imaging and the reason for and advantages of having overlapping color responses from imaging filters.

 

   Below of two CN searches that highlight some of their posts across a wide range of threads.

 

Frank's (freestar8n) writings regarding the Green Channel in astro-imaging:

 

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

 

 

   Hopefully these will give you some insights and possibly spark a discussion covering the areas you are interested in.

 

 

John

 

[Edits] Trying to get search terms better tuned...


Edited by jdupton, 22 June 2022 - 08:57 AM.

  • Dean J. likes this

#3 OhmEye

OhmEye

    Apollo

  • *****
  • Posts: 1,026
  • Joined: 15 Sep 2019
  • Loc: Western NY Southern Tier

Posted 22 June 2022 - 08:48 AM

The key factor for me in this periodic relative merits discussion is signal gathering efficiency, ie your "counting photons". There are definitely a lot of topics for RGB vs OSC discussions but the one I keep landing on myself is filter transmission efficiency. This is specific to hardware but in most cases I'm aware of the transmission for discrete external RGB filters is significantly higher than for most on-chip bayer matrix filters. All other gymnastics to be equal aside, in the end there simply is less signal per time reaching the OSC sensor. It's also very much worth being aware of potentially large differences in time efficiency between RGB and LRGB that also matter when comparing vs OSC.



#4 Eyeroll1952

Eyeroll1952

    Sputnik

  • *****
  • topic starter
  • Posts: 40
  • Joined: 05 Mar 2021
  • Loc: Berthoud, CO

Posted 22 June 2022 - 09:01 AM

Paul,

 

 

   If you have already read the other main CloudyNights Mono to OSC comparison threads, maybe your specific questions will be answered by looking at the responses from two specific CN members on those subjects. Frank (freestar8n), the author of MetaGuide and Mark (sharkmelley) have written a lot about both the importance of the Green channel in astro-imaging and the reason for and advantages of having overlapping color responses from imaging filters.

 

   Below of two CN searches that highlight some of their posts across a wide range of threads.

 

Frank's (freestar8n) writings regarding the Green Channel in astro-imaging:

 

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

 

 

   Hopefully these will give you some insights and possibly spark a discussion covering the areas you are interested in.

 

 

John

 

[Edits] Trying to get search terms better tuned...

Thanks so much for providing links. I will definitely check those out a bit later today.

 

Paul



#5 Eyeroll1952

Eyeroll1952

    Sputnik

  • *****
  • topic starter
  • Posts: 40
  • Joined: 05 Mar 2021
  • Loc: Berthoud, CO

Posted 22 June 2022 - 09:05 AM

The key factor for me in this periodic relative merits discussion is signal gathering efficiency, ie your "counting photons". There are definitely a lot of topics for RGB vs OSC discussions but the one I keep landing on myself is filter transmission efficiency. This is specific to hardware but in most cases I'm aware of the transmission for discrete external RGB filters is significantly higher than for most on-chip bayer matrix filters. All other gymnastics to be equal aside, in the end there simply is less signal per time reaching the OSC sensor. It's also very much worth being aware of potentially large differences in time efficiency between RGB and LRGB that also matter when comparing vs OSC.

Yes, I certainly didn't mean to discount the important of filter efficiencies. What has intrigued me since I started looking at spectral responses is the channel overlap. Since these Bayer filters are developed for general photography I now wonder whether that it at least partially deliberate or if it is just difficult to make such small (and thin) filters more selective.



#6 Eyeroll1952

Eyeroll1952

    Sputnik

  • *****
  • topic starter
  • Posts: 40
  • Joined: 05 Mar 2021
  • Loc: Berthoud, CO

Posted 24 June 2022 - 06:55 AM

Paul,

 

 

   If you have already read the other main CloudyNights Mono to OSC comparison threads, maybe your specific questions will be answered by looking at the responses from two specific CN members on those subjects. Frank (freestar8n), the author of MetaGuide and Mark (sharkmelley) have written a lot about both the importance of the Green channel in astro-imaging and the reason for and advantages of having overlapping color responses from imaging filters.

 

   Below of two CN searches that highlight some of their posts across a wide range of threads.

 

Frank's (freestar8n) writings regarding the Green Channel in astro-imaging:

 

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

Mark's (sharkmelly) writing regarding channel overlap in atro-imaging:

 

 

   Hopefully these will give you some insights and possibly spark a discussion covering the areas you are interested in.

 

 

John

 

[Edits] Trying to get search terms better tuned...

John,

 

Those links did indeed have interesting discussions but didn't necessarily resolve the questions I have so perhaps I just need to do more experiments myself. The issue of bleed between channels is clearly recognized but I was hoping to see some pictures to illustrate whether this has a significant influence on image quality - or perception really. 

 

I happened to have a 650 nm laser pointer available so diffused that beam and illuminated my ASI2600MC OSC sensor. This gave the histogram in the top part of the picture below and would be representative of what an OSC camera would see with Ha illumination. The blue and green channels pick up a fair amount of light. One then wonders whether blue and green channel contributions in images dominated by Ha are really mostly Ha. For such images one could subtract out the Ha contributions in blue and green and I may muck around with that.

 

Incidentally, the histograms shown are really broad in the red, in part due to non-uniform illumination, but also speckle due to the high coherence of the laser. To demonstrate that a broader band source would give narrower peaks I had a 525 nm LED around to produce the image at the bottom. Definitely narrower histogram peaks in this case.

 

Paul

 

Laser and LED spectra.jpg




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: Astrophotography, CMOS, DSO, Filters



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics