Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

2 Bin or not to Bin that is the question. LRGB


Earl

Recommended Posts

I am a big fan of binning colour data, especially at longer focal lengths, but, in reality even down to focal lengths of <500mm.  I don't see any point in collecting luminence data if you aren't going to bin the colour.  There are ways of creating a luminence channel by optimally combining the colour channels.  I believe Peter Shah does this and his images speak for themselves.

The problems people run into with binning is bloated stars.  Whilst, at shorter focal lengths, there will be a small loss of star resolution most star bloat comes from a relative increase in star exposure in the binned image compared with the unbinned.  Quartering the time of the binned image will give similar brightness to unbinned and in my experience doesn't bloat the stars.  Any loss of resolution will have disappeared by the time the image is shrunk for display on a computer monitor!  The benefit of binning is that you don't need as much overall exposure time for the colour and can spend this on gathering lots of extra luminence data.  Creating a luminence from colour channels is a good alternative but you still need extra exposure time or settle for extra noise because of the reduced light transmission through colour filters compared with luminence (works really well on narrow band images when you are unlikely to be using a clear lum filter).

There are imagers on SGL who are producing world class quality images without using binning but check out their overall exposure times.

It gets more complicated!  The actual benefit of binning isn't that it produces a brighter image for a given exposure because shot noise also increase so signal to shot noise isn't improved.  The benefit comes from a theortical quartering of read noise.  To cash in on this benefit it is necessary to have short exposures which would be read noise limited if unbinned.  

Link to comment
Share on other sites

Cheers

I did read that one also, im goign to ttry out 2x2 tonight if all goes well, just making my calibraion files now.

Earl,

You probably won't need calibration files for binned colour, unless you have massive vignetting. The RGB can easily stand quite a heavy blur as the detail is contained in the Lum layer.

Link to comment
Share on other sites

Earl,

You probably won't need calibration files for binned colour, unless you have massive vignetting. The RGB can easily stand quite a heavy blur as the detail is contained in the Lum layer.

Ah ok then :)

Link to comment
Share on other sites

I need to do some experimenting based on Martin's points. I have not reduced my binned exposure times sufficiently, perhaps, and I can see what he's saying. I don't think you can discuss this without getting into a discussion on what pixel scales represent a reasonable limit. I found that binning the Aitk 700O/Baby Q reduced was just not nice. Crude stars. Obviously coming down to an arcsec per pixel in a long FL rig through binning won't do anybody any harm. Tom also argues that binning at low elevations becomes problematic because your resolution is fuirther compromised by the seeing.

Two further arguments against binning;

1) In principle an unbinned RGB image should have the luminance equivalence of about 1/3 a real luminance of the same exposure. In reality I think it might be a bit less. You can add the synthtetic L from your RGB to your real L and put back some of the time lost by not binning. You cannot do this with a binned RGB because it doesn't have the resolution. I now routinely add synthetic L to L.

2) I make my starfields (without background nebulosity) out of RGB only. I get more colourful and smaller stars but binned RGB would be no good for this.

I never bin these days but will give it a go in the TEC on selected targets.

Olly

Link to comment
Share on other sites

Hi Olly

"binning the Aitk 700O/Baby Q reduced was just not nice."

Did you mean the Atik 4000? I just tried (last couple of nights) M13 with RGB binned 2x2 on my WO 98 with the Riccardi reducer and the Atik 4000. Haven't had a chance to process yet.

If you did mean the Atik 4000 I might re-shoot my RGB unbinned.

How is Cachou doing?

Cheers

Ian

Link to comment
Share on other sites

Hi Olly

"binning the Aitk 700O/Baby Q reduced was just not nice."

Did you mean the Atik 4000? I just tried (last couple of nights) M13 with RGB binned 2x2 on my WO 98 with the Riccardi reducer and the Atik 4000. Haven't had a chance to process yet.

If you did mean the Atik 4000 I might re-shoot my RGB unbinned.

How is Cachou doing?

Cheers

Ian

Hi Ian,

Not a clue where 700 came from and, yes, I meant 4000! Senior moment. (These moments generally last a week or two... :grin: )

Cachou is now tip top and running like a nutter on all four legs, even though one is not properly connected! She's fine and thanks for asking.

I would try your data and see. Martin's is a thoughtful point with regard to not over exposing while binning.

Very best,

Olly

Link to comment
Share on other sites

I binned the RGB 2X2 the night before last while going for the  very dim Tulip nebula. Using an Atik 314L and a 72 mm scope at F4.8, quite a few stars were well up to the 65000 ADU unit @ 400s exposure. I am not sure what this means as I had an Idas P1 filter on top becuase of the severe LP that I suffer from. Some of the smaller stars were really pixelated @ 100%.

A.G

Link to comment
Share on other sites

What constitutes 'short enought to be read noise limited'? I guess this depends on the chip and optical speed?

Have a look here http://www.ccdware.com/resources/

It doesn't cover binned subs but as a very crude measure you can divide your calculated sub exposure time by 4 for a binned image.  The less your read noise and the higher your sky glow the shorter your subs need to be.

Link to comment
Share on other sites

Olly

"Cachou is now tip top and running like a nutter on all four legs, even though one is not properly connected! She's fine and thanks for asking."

That's the best news I have had all year!

Love to Monique. And could you give Cachou a tickle behind the ears for me.

Cheers

ian

Link to comment
Share on other sites

If youre operating at a fairly short FL (500-600mm), then its possible to counter any problematic stars in the binned layer by meeting the scales halfway.

ie: Rescale your Lum layer by 25%, then rescale your binned layer by 150%.

That way, you havent lost much of the Lum scale, but you havent had to double the size of your binned data (which can often have bad effects on the stars).

Link to comment
Share on other sites

Binning is a means of resucing (random) noise in an image by combining nxn neighbouring pixels. thus 2x2 averages four pixes so will give a 2x reduction in random noise. The downside is that spatial resolution is lost. But if the image is being oversampled by the same amount then there is no loss of effective resolution. Even if the image is undersampled, then binning may be a good idea just to reduce the noise, albeit at expense of losing some resolution (you have to decide). 2x2 binning is equivalent to having to take 4x longer total exposure, 3x3 is 9x etc. The situation is more complex for OSC cameras with Bayer/matrix filters, but the principle is the same.

Patrick

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.