Celestia Forums

How "real" are the colors in these images? They aren??t accompagnied by any info on how they were made. Is this visible light? Does anyone know anything about it?

http://ciclops.org/view_media.php?id=8303
http://ciclops.org/view_media.php?id=2646
http://ciclops.org/view_media.php?id=574
http://ciclops.org/view_media.php?id=745
http://ciclops.org/view_media.php?id=4283 (big)
... Beware of the last one: it is in high resolution - 5.4 mb. Fantastic detail on the rings, btw),

-rthorvald

Three questions were asked by a polite, respected associate :

"How real are the colors in these images? "

The pictures have enhanced color that is intended to approximate the color seen by a human eye. When I look through a telescope at Saturn, it looks yellow, like the over-all color shown in the first photo. Looking through a 4 inch to 8 inch telescope at Saturn is very memorable. The beauty comes through in live viewing of a tiny image better than these big, detailed, flat 2 dimensional pictures. Saturn has a pale yellow color in general, when viewed through a small telescope. Naked-eye viewing also shows the pale yellow color of Saturn, "the bringer of old age" as the ancients called it.

To see un-enhanced color look at this :
http://antwrp.gsfc.nasa.gov/apod/ap000129.html

"Is this visible light?" No.

Images taken on Oct. 11, 2005, with blue, green and infrared (centered at 752 nanometers) spectral filters were used to create this color view, which approximates the scene as it would appear to the human eye. The images were obtained with the Cassini spacecraft wide-angle camera at a distance of approximately 39,000 kilometers (24,200 miles) from Dione and at a Sun-Dione-spacecraft, or phase, angle of 22 degrees. The image scale is about 2 kilometers (1 mile) per pixel.

see http://spacenews.dancebeat.info/article ... _ice_world


"Does anyone know anything about it? "

Yes : search engine : google images dione

Images taken on Oct. 11, 2005, with blue, green and infrared (centered at 752 nanometers) spectral filters were used to create this color view

We can't see in the IR, so the answer has to be "no", although it's probably close. I suspect the IR values aren't very far from what a red channel would provide. Also, the people who created the resulting images may have adjusted the color balance to match whatever they've defined to be the visual color distribution. *shrug*

GlobeMaker wrote:The pictures have enhanced color that is intended to approximate the color seen by a human eye [...]
To see un-enhanced color look at this :
http://antwrp.gsfc.nasa.gov/apod/ap000129.html

selden wrote:I suspect the IR values aren't very far from what a red channel would provide. Also, the people who created the resulting images may have adjusted the color balance to match whatever they've defined to be the visual color distribution. *shrug*

Thanks, boh, for your replies (and for the link, GlobeMaker).

So, can i assume that these images are the ciclops??teams effort to present them in as close to "natural" colors as possible (what i would see if i was there)? If so, the Celestia default texture is a bit too saturated?

And, looking at the link GlobeMaker presented, colors should perhaps be toned down even more than the ciclops images?

-rthorvald

Hello rthorvald,

In my opinion, yes, the color should be toned down. But the human perception of color is not a fixed fact. See :

http://en.wikipedia.org/wiki/Color

Edwin Land in the 1970s published notes on this perception. (He worked on the Polaroid Land Camera).

The colorful pictures of Saturn that you showed are influenced by wishful thinking, illumination, and adjacent colors. For example, if you print out the pictures on paper, you can look at them with the following lighting and adjacencies :

moonlight in a grassy field

moonlight on a red carpet

sunlight against yellow or blue backgrounds

florescent lights

incandescent light.

red lights and white lights together

after a night's rest, the first thing you look at is a Saturn picture by dawn light

after staring at a CRT monitor, look at the paper picture


Please perform the following easy test of your color perception :
Look at a sunset for 30 seconds.
Rotate your head 20 degrees clockwise.
Notice how the left half of the sunset has a different color than the right half. Your eyes and brain have adapted to the sunset in a way that changed the colors! The same is true for Saturn : an astronaut over Saturn will adapt to the colors and change the colors automatically and uncontrollably.

The colors that a human sees depends on these factors. There are psychological factors, as the Wikipedia article spells out. The bottom line : use a telescope to see the true color of Saturn. Then you be the judge.

That unenhanced photo is beautiful, and I'd forgotten to check out that site for the new ones in a while.
Thanks for the reminder!

GlobeMaker wrote:The colors that a human sees depends on these
factors. There are psychological factors, as the Wikipedia article spells
out. The bottom line : use a telescope to see the true color of Saturn.
Then you be the judge.

Thank you for your reply. As a designer (and professional illustrator) i
am certainly aware of how subjective colors are and behave. In this
instance i am trying to learn a bit more about Saturn vs the Cassini
imagery from Ciclops... Thanks for the links; the Saturn image was very
interesting.

Now, on to something else: i am also trying to make some sense of the
scale of images from the Ciclops site. Below i have placed a number of
Cassini images on Bj?¶rn Jonssons Saturn map.

This type of work is completely new to me. Of course there are lots of
distortion in these shots, but accounting for that, what do you guys think?
Is this reasonable?


(... This might belong in the textures forum, but i have more need of
opinions from astronomers than from texture artists here...)
-rthorvald

Oh, yes: the purpose of this exercise is to paint a more detailed Saturn-like
map. I don??t have enough data to make an exact Saturn (for example,
there is almost nothing on the nothern hemisphere), but there are material
to do something of a reasonable impression. Here is a quick test (no
painting here, just the merging of a bunch of Cassini images onto Bj?¶rn??s
Saturn map):



Ideas on how to proceed - and suggestions - are most welcome.

-rthorvald

Hello rthorvald,

You asked for suggestions. Please consider the relative detailing of
Saturn versus Jupiter. Saturn is bland compared to Jupiter. Many
closeup pictures are available of both planets, and Saturn has a
lower contrast of dark and light than Jupiter. It is not just the color
of Saturn which is pale and relatively featureless compared to
Jupiter. Even black and white photos show the great difference in the
light area versus dark area contrast. So please be cautious about
producing Saturnian details that thrill the viewer with unrealistic contrast.

My suggestion, is to produce five cloud layers, each with increasing
contrast and each at a different height above the core.
When seen from far away, all is bland. But as a visitor
travels down below the top cloud layer, more contrast is visible.
Each of five cloud layers can have increasing constrast and color
saturation. The bottom layer can be a rocky core with interior as well as
exterior normals, so views are seen at any position within tha planet.
Please look at this model of Jupiter's cloud decks, where the vewer is
below the top cloud deck :

http://antwrp.gsfc.nasa.gov/apod/ap971030.html

In that essay, it discusses the "dark blue area, a relatively clear,
dry region similar to the site where Galileo's atmospheric probe
made the first entry into a gas giant planet's atmosphere on
December 7th, 1995".

Please use the following suggestion to provide clear zones in the
atmosphere that go to unexpected depths through four layers:

There is a menu choice to show clouds and atmosphere. Your five
layers can be linked to those choices. For example, when all clouds
and atmosphere are shown, all is bland. When atmosphere is unselected,
contrast is enhanced to simulate infra-red visiblity. When clouds are
unselected, contrast is enhanced to simulate UV visibility. When both
clouds and atmosphere are deselected, the very detailed, highest contrast
Saturn with enhanced colors is shown. In this way, we get the best
of both worlds: realistic and enhanced.

Cloud bands can rotate and counter-rotate.
The "Great Yellow Spot" on Saturn can be produced as a false moon.
So it is written, so let it be.

GM,

Unfortunately, Celestia doesn't work that way.
The Show Atmosphere option causes Celestia to draw a hazy ring around the planet's rim. It can't be used to select among different image textures.

However, the lores/medres/highres option (type "r" or "R") can be used to change both surface textures and cloud textures. And so can AltSurface, although it's clumsier.

selden wrote:GM, Unfortunately, Celestia doesn't work that way.
The Show Atmosphere option causes Celestia to draw a hazy ring around
the planet's rim.

But the basic idea can be used differently... By having no detail at the 1k
level and more detail at 8k virtual texture...

I am aware that Saturn is very bland compared to Jupiter, but it is logical
that *some* detail will be visible close up. Most of the Ciclops images
shows detail in the infrared, but my plan was to use these as a roadmap
for painting the entire globe, then merge it into a regular Saturn map with
very high level of transparency - the result will just hint at the detail, not
show it very well. After that, applying some blur effects to the lower-res
VT levels showuld produce an effect similar to what GlobeMaker
suggested: one will have to be very close to the planet to see the cloud
details.

If anyone is aware of any work being done on Saturn mapping
elsewhere, please post links...

-rthorvald