#1  
Old 28-03-2010, 02:28 AM
Dietmar's Avatar
Dietmar
anatomic astronomer

Dietmar is offline
 
Join Date: Jul 2007
Location: Linz, Austria
Posts: 372
about the colors in astro-imaging

hi guys,

to achieve relastic colors is what we are eager to obtain in all our astro-imaging efforts, isn't it? however, the question would be: what is realistic? what is objective ?
and what is feasible in color-imaging?
I have witnessed many discussions about the "true colors in LRGB- and DSLR-astrophotography" that repeatedly have taken place all over the globe in different forums at several occasions and frequently yield a result which clearly misses the punch-line of human color-sensation.
You might have attended some astro-imaging conference where a couple of well renowned imagers have presented their photos of maybe the same object and claimed to have gotten the "right" colors. And has there not been a huge difference in the appearance of such photos? Naturally! Because the the most important factor in color-composition is us.
here are some thoughts about colors in astronomical imaging:
a bit of physiological and philosophical background from my standpoint as physician and photographer.

http://www.stargazer-observatory.com/art-of-seeing.html
Reply With Quote
  #2  
Old 28-03-2010, 03:11 AM
Ritchie-l's Avatar
Ritchie-l (Ritchie)
Registered User

Ritchie-l is offline
 
Join Date: Jan 2010
Location: Craigieburn, Vic
Posts: 40
funny you should say that.
I have looked at the Orion nebula for instance, through my 8"
and to me it is shades of grey.
I took a pic with my canon a590is and the clouds are a blueish tint.
I used the webcam and they are greyish.
so i bought a CCD camera.

all because i thought they were red.
i got this idea from a pic i saw in an astronomy book.
so i didn't trust my eyes as they are aging.
I didn't trust the webcam cause I didn't pay to much for it.
I didn't trust the canon as the primary lens is removed.
all because of an impression that said it was red.

so i am with you on this one.
what is wrong with what the eye see's.
Ritchie
Reply With Quote
  #3  
Old 28-03-2010, 09:27 AM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Great read Dietmar. THanks for the link.
Reply With Quote
  #4  
Old 28-03-2010, 09:45 AM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,280
Breathtaking!

Thank you Dietmar!

Steve
Reply With Quote
  #5  
Old 29-03-2010, 02:26 AM
mjc's Avatar
mjc (Mark)
Registered User

mjc is offline
 
Join Date: Mar 2010
Location: Dublin, Ireland
Posts: 52
Dietmar

Thankyou for your post - in fact it was your post that encouraged me to bite the bullet and join this forum today in recognition of the quality of the forum (whenever I've been searching for answers I often find myself at an iceinspace thread).

I enjoyed the post. First I'd like to point out that in the text immediately above the graph you accidently (I think) swap the terms cones and rods.

I think its a wonderful article that you should develop it further over time.

Material I'd like to suggest for inclusion is the relationship between the physiology of the eye and the decision by Bayer to double-up the green sensor in the Bayer matrix - which I seem to recall was to match the greater acuity of the eye in green.

The circular ring of rods is new to me and is quite thought provoking.
I am trying to work out what benefit the ring structure would have and can only (weakly) associate it with the saccades (rapid movement movement) of the eye which (at least to some degree) allows it to identify boundaries and thus achieve data-compression for transmission to the brain over a less dense nerve bundle which you describe.

The eye only has a small aperture - its diffraction limited resolution has to be poor. It would be enlightening to find what the eye's "pixel scale" (at fovea and elsewhere) is and to determine if the saccades contributed to dithering/resampling and whether there are techniques that we can learn and transfer to our astroimaging process.

You refer to true colour and false colour, may I also introduce the concept of "representative colour". This is where we match the longest wavelength image to red, the shortest wavelength image to blue, and whatever we have in the middle to green. We are true to the order of the wavelengths but they need not be the familiar RGB that the eye can see (eg we can use IR or even radio for red and UV for Blue).

I gleamed that snippet from the following which is worth a read in general:
http://www.spacetelescope.org/projec...or/improc.html

Mark
Reply With Quote
  #6  
Old 29-03-2010, 08:08 AM
tonybarry's Avatar
tonybarry (Tony)
Registered User

tonybarry is offline
 
Join Date: Aug 2007
Location: Penrith, Sydney
Posts: 556
Quote:
to achieve relastic colors is what we are eager to obtain in all our astro-imaging efforts, isn't it? however, the question would be: what is realistic? what is objective ?
Hello Dietz,

You make a very good point here, one which has occupied my thinking since I obtained a monochrome camera for astro imaging and had to assemble an RGB image from three individual colour representations.

I recall asking the question here - how does one do a "colour balance"? when it's not possible to see the object one is imaging (or see it only very faintly?)

The best answer I got was to do a colour balance on a star which has a known spectral hue (e.g. Betelgeuse for reds, Rigel for blue) ... which does offer some kind of a reference white (or colour).

It is partially for this reason that I think the OSC (one-shot-colour) cameras for astro work may be more reasonable items to use, even though the Bayer Mask does cut down the light intensity considerably.

I hope you can continue your thinking on the subject, and look forward to your thoughts.

Regards,
Tony Barry
Reply With Quote
  #7  
Old 29-03-2010, 09:39 AM
Barrykgerdes
Registered User

Barrykgerdes is offline
 
Join Date: Feb 2007
Location: Beaumont Hills NSW
Posts: 2,900
The display of colour in astronomical photos is always going to be false because we use our eyes to detect the different frequencies of the emmissions. Colour, like light and dark is how our brains interpret the signals from the receiving aerials in the eye. Who's to say that what I perceive is the same as what others see.

The more sensitive detectors accumulate the luminance of the object which is all we see till the intensity of the accumulated light allows sufficient data to fall on the specific frequency detectors that can pass the color information to the brain.

We ascribe colours to our astronomical photos to the spectrum colour of the material that is emitting or reflecting the light according to how photo sensitive receptors measure it according to the radiated frequencies and then display it usually with a computer grahics program set up to what the eye likes to see.

The most blatent of falsely displayed colours is now done with what we refer to as narrow band photography where our cameras detect the hydrogen alpha components in the sub red area of the spectrum and then display it as a false colour or accentuated luminance. I must admit though that the result is usually very pleasing to the eye and this is really what we like.

So who really cares about the colour. The scientific measurements of colour emissions are used for a multitude of purposes but what we like to see is purely up to our own preferences and usually does not have any relationship to actual colours.

I know I prefer the colours in astro photos to have less colour saturation than is commonly shown. But that is only my preference.

Barry
Reply With Quote
  #8  
Old 29-03-2010, 09:44 PM
tonybarry's Avatar
tonybarry (Tony)
Registered User

tonybarry is offline
 
Join Date: Aug 2007
Location: Penrith, Sydney
Posts: 556
Barry, I appreciate your thoughts.

My take is that, if I were close enough to M42 to appreciate the colour (without suffering a bad dose of high energy particles and radiation) then I would like my astro images to show me that colour I'd see IF I were closer.

I do care about the colour; the false colour images are there for a purpose and they present their data well; but for aesthetic appreciation of an object, the "actual" colour is what I'd like. To me there is some art in the images I see here on Ice In Space, and I acknowledge the skill and the artistry, and the images can be both captivating and useful for science. But for my own appreciation of the object - let me see it as it would be if I were closer.

I do appreciate that in a number of cases, the overwhelming light from a primary would absolutely cook out any chance I'd have of seeing the surrounding glow. Turning the glare down with a digital version of sunglasses is OK. But the minimum is good.

Regards,
Tony Barry
Reply With Quote
  #9  
Old 29-03-2010, 10:02 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,280
Quote:
Originally Posted by tonybarry View Post
I recall asking the question here - how does one do a "colour balance"? when it's not possible to see the object one is imaging (or see it only very faintly?)

The best answer I got was to do a colour balance on a star which has a known spectral hue (e.g. Betelgeuse for reds, Rigel for blue) ... which does offer some kind of a reference white (or colour).

It is partially for this reason that I think the OSC (one-shot-colour) cameras for astro work may be more reasonable items to use, even though the Bayer Mask does cut down the light intensity considerably.

I hope you can continue your thinking on the subject, and look forward to your thoughts.

Regards,
Tony Barry
Tony,

it's best to try and calibrate your OSC and/ or mono cameras+RGB filters
on G2V stars. This is the only way you can get the RGB mix even
close. Meade filters will give different G2V calibration settings to
Astronomiks etc. Because CCDS are much more linear than emulsion
film, once this calibration factor is known and set, then your colour
balance should be a reasonably close approximation to true colours.
Another annoying thing is atmospheric extinction. If you take your
RGB set low in the sky, the blues will be affected most by this. FWIW.


Steve
Reply With Quote
  #10  
Old 30-03-2010, 07:58 AM
bojan's Avatar
bojan
amateur

bojan is offline
 
Join Date: Jul 2006
Location: Mt Waverley, VIC
Posts: 6,932
Quote:
Originally Posted by tonybarry View Post
.....if I were close enough to M42 to appreciate the colour....
Tony, you can go as close to M42 as you like, but the surface brightness of the nebula (and your colour perception) would remain exactly the same as it is from where you are now.

The only thing that would help you to see "real" colours is aperture (and you can have it here on Earth )
But then, this is also a form of illusion.. because irises of our eyes are 8mm maximum in diameter and they can collect only this much light... so any aid here is actually falsifying the reality.

Last edited by bojan; 30-03-2010 at 08:44 AM.
Reply With Quote
  #11  
Old 30-03-2010, 08:34 AM
avandonk's Avatar
avandonk
avandonk

avandonk is offline
 
Join Date: Aug 2005
Location: Melbourne
Posts: 4,786
Colour does not exist! There is no intrinsic quality of any spectral interval of EM radiation that has a colour.

Our human colour vision is the resulting stimulus of different spectral luminances on the RGB sensitive cones in the eye. The brain interprets these differences in spectral stimulus as 'colour'. Dietmars explanation of course is far fuller.

Human females can see far more hues than human males as the genes for colour vision are on the x chromasome.

To show how important the relative spectral intensities are and the environment they are in, google Land Effect. Edwin Land was the inventor of instant photography (Polaroid) and some others.

Here is a very good explanation of the Land Effect.

http://www.aw3rd.us/scief/colorviz.htm

This effect also works with two filters of the same colour but different spectral content. The effect is weaker but still there.

Most people cannot believe their eyes and suspect trickery when this is done with two slide projectors.

It is also the basis for some colour optical illusions.

Bert

Last edited by avandonk; 30-03-2010 at 08:56 AM.
Reply With Quote
  #12  
Old 30-03-2010, 08:51 AM
bojan's Avatar
bojan
amateur

bojan is offline
 
Join Date: Jul 2006
Location: Mt Waverley, VIC
Posts: 6,932
This effect works because white light contains full spectra, and superimposed one colour adds to impalance.. the rest is psycho-physiological effect (most likely pre-processing in the eye?).
Also I think couple of equations may be useful here (to show how colour imbalance produced with two projectors indicate the "true" colours, after re-balancing in the brain)
Reply With Quote
  #13  
Old 30-03-2010, 09:12 AM
avandonk's Avatar
avandonk
avandonk

avandonk is offline
 
Join Date: Aug 2005
Location: Melbourne
Posts: 4,786
That is quite correct Bojan. It works because the two b&w slides are modulating the relative luminances of the full spectrum of the projector globe. The red filter or any other then further modulates the relative spectral luminances of each part of the image.

I first saw this effect demonstrated in 1969 at Kodak Research Laboratories. You have to know a lot about the human visual system and different lighting conditions to make colour film.

Remember Daylight and Tungsten type film? Or filters that could get a white colour balance under either. Those were the days!

Bert

Last edited by avandonk; 30-03-2010 at 09:26 AM.
Reply With Quote
  #14  
Old 30-03-2010, 09:15 AM
bojan's Avatar
bojan
amateur

bojan is offline
 
Join Date: Jul 2006
Location: Mt Waverley, VIC
Posts: 6,932
Yes
We were also much younger then
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 08:35 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement