| |
Main
Date: 08 May 2007 16:24:36
From: canopus56
Subject: How do colored planetary filters work?
|
I have been doing some reading, trying to get a handle on the effects
of color planetary filters by planet. I believe I understand it but
wanted to get some confirmation not being up on color theory.
I have not learned modern color theory but have a working
understanding of Newton's outdated color chart:
http://en.wikipedia.org/wiki/Image:2007_charles_blanc_etoile_des_couleurs_1867.PNG
Primary and secondary opposite colors mix to produce a colorless
white, black or grey. E.g. yellow and violet or red and green.
Planetary color filters are transulcent. Like opaque objects, they
obtain their color by absorbing colors other than their apparent color
and reflecting or transmitting the apparent color. A red filter
appears red because it absorbs reflects red and absorbs other colors.
If a primary color is passed by a colored filter, then the opposite
colors on the color wheel are suppressed. For a red filter, green and
blue are reduced in brightness. This reduction in brightnesses
reduces apparent color mixing to a grey hue as perceived in the eye.
Paradoxically, this increases contrast of the opposite colors and the
eye's ability to perceive details of features with the opposite colors
that are now not completely mixed with its opposite dominent color. A
red filter increases the detail seen in features that are colored blue
and green.
So the product inserts for let's say a Meade 25A dark Red filter
states on Jupiter, that it "increases visiblity of Great Red Spot.
Increases contrast of belt details." Presumably this means it passes
some red but suppresses more blue and green, allowing more contrast in
the blue and green details to be seen.
Conversely, for a Meade 58 Green filter, the inserts describes how it
on Jupiter it "increases contrast in lighter parts of the Jupiter disk
by supressing red and blue toned structures. Strongly rejects blue and
red tones. Use with 8" of apeture or greater."
The recommendation to use larger apetures, e.g. 8 or 10 inches,
depends on the percent of light that a filter transmits. High-
transmission unsaturated filters can be used with smaller apetures.
Low-transmission heavily-saturated filters need larger apetures to
compensate for the loss of light.
An analogous process occurs in modern digital astrophotography
processing. A single color channel is suppressed - resulting in
incomplete color mixing. So, instead of seeing featureless grey, you
can see detail in the opposite partially mixed colors.
Does this sound about right?
How would you describe the operation of color planetary filters in
terms of modern color theory with its three primary additive colors
and three primary subtractive colors?
See http://en.wikipedia.org/wiki/Color_theory
- Thanks in advance - Kurt
Primary and secondary opposite colors mix to produce a colorless
white, black or grey. E.g. yellow and violet or red and green.
Planetary color filters are transulcent. Like opaque objects, they
obtain their color by absorbing colors other than their apparent color
and reflecting or transmitting the apparent color. A red filter
appears red because it absorbs reflects red and absorbs other colors.
If a primary color is passed by a colored filter, then the opposite
colors on the color wheel are suppressed. For a red filter, green and
blue are reduced in brightness. This reduction in brightnesses
reduces apparent color mixing to a grey hue as perceived in the
eye.
Paradoxically, this increases contrast of the opposite colors and the
eye's ability to perceive details of features with the opposite
colors. A red filter increases the detail seen in features that are
colored blue and green.
So the product inserts for let's say a Meade 25A dark Red filter
states on Jupiter, that it "increases visiblity of Great Red Spot.
Increases contrast of belt details." Presumably this means it passes
some red but suppresses more blue and green, allowing more contrast in
the blue and green details to be seen.
Conversely, for a Meade 58 Green filter, the inserts describes how it
on Jupiter it "increases contrast in lighter parts of the Jupiter disk
by supressing red and blue toned structures. Strongly rejects blue and
red tones. Use with 8" of apeture or greater."
The recommendation to use larger apetures, e.g. 8 or 10 inches,
depends on the percent of light that a filter transmits. High-
transmission unsaturated filters can be used with smaller apetures.
Low-transmission heavily-saturated filters need larger apetures to
compensate for the loss of light.
Does this sound about right? How would you describe the operation of
color planetary filters in terms of modern color theory with its three
primary additive colors and three primary subtractive colors?
See http://en.wikipedia.org/wiki/Color_theory
- Thanks in advance - Kurt
|
|
| |
Date: 10 May 2007 12:28:03
From: canopus56
Subject: Re: How do colored planetary filters work?
|
On May 9, 3:42 pm, pau...@saaf.se (Paul Schlyter) wrote:
> Color is the sensation produced in the eye when we see
> things around us. It's wrong to claim that the color
> "is nothing to do with the light coming off the object"
> - that light matters a lot for the color we perceive
> when we see the object.
Paul, thank you for the notes on color perception. After further
reading, I'm thinking along those lines and your comments were helpful
getting me pointed the right way.
Here's the short rule-of-thumb for colored filters that I was looking
for:
"What you see is that a filter lightens tones of objects colored
similarly to the filter and darkens tones of objects colored
dissimilarly."
Kodak. 1998. Using Filters (Kodak Workshop Series) (The Kodak Workshop
Series) (Paperback).
So, a red filter on Mars lightens the reds and darkens the blue polar
ice caps and the blue-green maria.
This is "lightening" and "darkening" in terms of psychological color
perception related to physical contrast changes from relative
reductions in luminance. It is not actual "lightening" or "darkening"
in terms of luminance.
For other lurker's benefit, here is a short experiment that I did with
my filters yesterday that clarified in my own mind on how the
mechanism of action of color planetary filters work. It appears the
similar-color-lightening and opposite-color-darkening effects are a
basic photographic filter principle used by terresterial
photographers. Maybe some of you camera buffs can chime in.
Take your box of astronomy filters outside before sundown. Look for
an neighborhood area that has lots of different colored flowers in
it. Spring is good time of the year to do this. Fortunately, my
neighbor has a green thumb and I had a good selection right outside my
door.
Look at a red flower through your dark red #25 filter. The flower
will appear white while the green leaves will be several tones
darker.
Through a #15 Deep Yellow, a yellow flower looks white while the green
leaves will appear darker.
Look at a green leafy tree through your #58 Green filter - it will
look lighter, not darker.
Use a #80 Blue on a blue flower - it will appear light and white while
the green will appear nearly black.
My Purple filter makes green trees appear jet black, blue flowers
disappear; yellows remain visible.
When applied to planets, these terresterial effects probably are all
dependent on getting enough light into the eye to trigger your color
vision - hence the need for big light-bucket apeture.
This underscores one of the points omitted from many manufacturer
product inserts - low-transmission percentage filters only work with
larger apetures (8", 10" 11" or above). This is because enough light
has to reach the eye in order to trigger color vision perception _when
the reduced level of light is viewed through filter_.
Paul also wrote:
> Did you ever use a spectrophotometer to actually obtain a
> spectrum of the light reflected from a surface which has
> been painted with your mixture of blue+yellow paint?
> If you did, you might be surprised to find that the
> reflected light *does* peak in the green!
Yes, I do have a hand-held Star spectrometer. But what about the
inverse case - putting a blue filter in front of a hand-held classroom
spectrometer looking at green light?
Thanks again.
- Kurt
|
| | |
Date: 11 May 2007 07:42:23
From: Paul Schlyter
Subject: Re: How do colored planetary filters work?
|
In article <1178825283.356086.301120@l77g2000hsb.googlegroups.com >,
canopus56 <canopus56@yahoo.com > wrote:
> Paul also wrote:
>> Did you ever use a spectrophotometer to actually obtain a
>> spectrum of the light reflected from a surface which has
>> been painted with your mixture of blue+yellow paint?
>> If you did, you might be surprised to find that the
>> reflected light *does* peak in the green!
>
> Yes, I do have a hand-held Star spectrometer. But what about the
> inverse case - putting a blue filter in front of a hand-held classroom
> spectrometer looking at green light?
That's just another case of subtractive color mixing.
Let's assume the green light is produced by putting a green filter in
front of a source of white light. If you put a blue filter on top of
that, what you see is simply the light passing through both filters.
And how much light passes through both filters depend on how much the
transmission curves (as a function of wavelength) overlap. If the two
filters are sufficiently narrow banded, there may be no overlap, and
in such a case no light will pass through both filters. Usually there
is some overlap though, but the brightness and the color of the light
passing through both filters will depend a lot on the details of this
overlap. As a result, if you have two pairs of green and bloe
filters, and the two blue filters look quite similar in brightness and
color, and the two green filters also look similar, the light passing
through green1+blue1 may look quite different from the light passing
through green2+blue2.
--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stockholm dot bostream dot se
WWW: http://stjarnhimlen.se/
|
| |
Date: 09 May 2007 17:15:32
From: Andrew Smallshaw
Subject: Re: How do colored planetary filters work?
|
On 2007-05-08, canopus56 <canopus56@yahoo.com > wrote:
>
> How would you describe the operation of color planetary filters in
> terms of modern color theory with its three primary additive colors
> and three primary subtractive colors?
> See http://en.wikipedia.org/wiki/Color_theory
Put simply they're different areas. Colour mixing relates to the
behaviour of the human eye and other detection equipment such as
colour CCDs (not sure about film) that emulate it.
If you mix yellow paint and blue paint an artist will tell you that
you end up with green paint. In a way he's right - the paint
certainly _looks_ green. However, in fact the paint is not green
but blue and yellow _at_the_same_time_. Your eye cannot cope with
an object being two colours simultaneously so resolves the object
into green. This is done _in_the_eye_, it's nothing to do with
the light coming off the object.
To adequately explain the effects of coloured filters it's necessary
not to consider colour at a point on the line between red and
violet, or even as a point in a sphere or cube, but as a graph of
frequency (from red to violet) versus intensity.
A green colour may have a peak in the green region and relatively
little at other frequencies, or, as in the blue-and-yellow paint
example, peaks in both the blue and yellow frequencies and relatively
little in the green region. The two colours would both _look_
green even though they have very different looking graphs.
--
Andrew Smallshaw
andrews@sdf.lonestar.org
|
| | |
Date: 09 May 2007 21:42:23
From: Paul Schlyter
Subject: Re: How do colored planetary filters work?
|
In article <slrnf440de.ht5.andrews@sdf.lonestar.org >,
Andrew Smallshaw <andrews@sdf.lonestar.org > wrote:
> On 2007-05-08, canopus56 <canopus56@yahoo.com> wrote:
>
>> How would you describe the operation of color planetary filters in
>> terms of modern color theory with its three primary additive colors
>> and three primary subtractive colors?
>> See http://en.wikipedia.org/wiki/Color_theory
>
> Put simply they're different areas. Colour mixing relates to the
> behaviour of the human eye and other detection equipment such as
> colour CCDs (not sure about film) that emulate it.
Color film surely tries to emulate the eye too - or else, the color
reproduction won't be any good.
> If you mix yellow paint and blue paint an artist will tell you that
> you end up with green paint.
Actually, if the two paints are a perfect blue and perfect yellow
(in the sense of additive primary colors, where yellow reflects both
red and green), you'll end up with black or some flavor of dark gray.
If you want "blue" and "yellow" paint to mix into green, the "blue"
must reflect a lot of green as well. Technically, such a blur color
is called "cyan" - that's the term used in the color print industry.
> In a way he's right - the paint certainly _looks_ green. However,
> in fact the paint is not green but blue and yellow _at_the_same_time_.
> Your eye cannot cope with an object being two colours simultaneously
> so resolves the object into green. This is done _in_the_eye_, it's
> nothing to do with the light coming off the object.
You're here confusing wavelength bands with color. Color is the
sensation produced in the eye when we see things around us. It's
wrong to claim that the color "is nothing to do with the light coming
off the object" - that light matters a lot for the color we perceive
when we see the object. But the color depends on other things too,
such as the wavelength mix of the light illuminating the object,
the light coming off other objects nearby which we see at the same
time, and even the colors we've seen recently. Color vision is
an incredibly complex subject, really.
> To adequately explain the effects of coloured filters it's necessary
> not to consider colour at a point on the line between red and
> violet, or even as a point in a sphere or cube, but as a graph of
> frequency (from red to violet) versus intensity.
>
> A green colour may have a peak in the green region and relatively
> little at other frequencies, or, as in the blue-and-yellow paint
> example, peaks in both the blue and yellow frequencies and relatively
> little in the green region.
Here you're confusing additive and subtractive color mixing....
If you apply no paint to a white surface, it'll naturally look white.
The more you paint it with various colors, the darker it'll look, and
finally it ends up loking more or less black. This is called
subtractive color mixing: we start with white, and subtract various
wavelength bands until we end up with the color we want. Subtractive
color mixing is used in photographs (slides as well as prints), other
prints, and painted surfaces.
If you instead start with a red + green + blue lamp illuminating a
white surface, and if the lamps are all turned off initially, the
surface will look black. By turning on the lamps in different
proportions, you can make the white surface appear to have any color
you like. If you light the red and the green lamps, the surface will
look yellow. If you then also turn on the blue lamp, the surface will
look white. This is called additive color mixing: we start with
black, and add various wavelength bands until we end up with the color
we want. Additive color mixing is used on TV and video screens: take
a magnifying glass and look closely at a yellow sorface on your
computer video monitor - you may be surprised when you see no yellow
dots there, only red and green dots.
Your blue-and-yellow paint example works like this: the yellow paint
absorbs blue, while the cyan (i.e. "blue") paint absorbs red. They
both reflect green. Now, if you paint a white surface with yellow,
it'll absorb blue but reflect red and green - and we see that as
yellow. Now, mix in blue paint, and the red will be absorbed as well,
and only the green will be reflected - the painted surface now looks
green, since it reflects mostly green. It does not reflect a lot
of blue and yellow as you suggested (except the part of the yellow
which resides in the green wavelength band of course) - remember that
this is subtractive, not additive, color mixing!
> The two colours would both _look_ green even though they have very
> different looking graphs.
Did you ever use a spectrophotometer to actually obtain a spectrum
of the light reflected from a surface which has been painted with
your mixture of blue+yellow paint? If you did, you might be surprised
to find that the reflected light *does* peak in the green! But don't
just trust my word here -- do the experiment!!!! If you don't have
access to any spectrophotometer, you can make a simplified experiment:
photograph that painted surface with a digital camera, bring the image
into a suitable image editor program, and check the RGB values of
the pixels within that green area. Yes, the camera will here act
as a (very coarse) spectrophotometer.
--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stockholm dot bostream dot se
WWW: http://stjarnhimlen.se/
|
|
