IPB

Welcome Guest ( Log In | Register )

Processing VIMS cubes, An attempt at "true" color
ugordan
post Sep 10 2006, 07:51 PM
Post #1


Senior Member
****

Group: Members
Posts: 3648
Joined: 1-October 05
From: Croatia
Member No.: 523



Right, a suggestion I made here in another topic made me wonder why not try that myself. A bunch of data was sitting on the PDS, after all. After a hassle figuring out just how the image cubes are organized and trying to read them, finally I was able to produce some results. This is all very rough work, can be considered first-iteration only and not particularly accurate.
Basically, I used the cubes to extract the visible spectrum in the 380-780 nanometer range which was then input to color matching code I found here by Andrew T. Young.
The code integrates over 40 10-nm steps to produce CIE XYZ color components. I then converted these to RGB values.

I'm aware of at least three inaccuracies in my code as of yet: one is the above sampled code apparently uses Illuminant C as the light source, not true solar spectra so the color turns out bluish (has a temp. of 9300 K instead of 6500 K, AFAIK). I tried to compensate at the moment by changing the final RGB white balance, but this is probably an inaccurate way to go. Another inaccuracy is I don't do bias removal from the cubes. This likely affects the outcome. Also, I don't use the precise wavelengths the code requires, but use the closest one in the cube. I intend to fix this by interpolating between nearest wavelengths.

All images are enlarged 4x.

Attached Image
Attached Image
Attached Image

The leftmost image is a 4-cube mosaic. The colors in all four frames turned out identical which gives me at least some confidence. The image in the middle shows Dione's disc creeping in front of Saturn. Dione's disc appears elongated probably because as the lines were readout, it moved considerably in its orbit. The rightmost image shows a very overexposed Saturn image, the part below the ring shadows got overexposed. From what I've seen browsing through the PDS, a lot of the cubes are badly overexposed at some wavelengths.

Here's a couple of Jupiter images. I'm not very satisfied with them as they seem to look somewhat greenish, but overall the color looks believeable:
Attached Image

Attached Image


Lastly, two Titan composites. They turned out way more reddish than I thought they would.
Attached Image
Attached Image


It'll be interesting to see how much the results will change once I do a more proper processing pipeline working.


--------------------
Go to the top of the page
 
+Quote Post
 
Start new topic
Replies
slinted
post Sep 21 2006, 08:52 PM
Post #2


Member
***

Group: Admin
Posts: 468
Joined: 11-February 04
From: USA
Member No.: 21



Good points made by all,

Part of my confusion on this might be coming from a misunderstanding about what the dataset is providing. Are the VIMS cubes in spectral power, or reflectance? (I appologize if I'm using these terms incorrectly) Has the illuminant spectrum already been removed?

If the VIMS data is in spectral power, then I would say that chromatic adaptation is still necessary since, as Don said, D65 isn't solar spectrum. It is what we see here on Earth.

If it is a reflectance dataset, then wouldn't that spectrum have to be multiplied by the spectrum of the D65 illuminant before being passed to the CIE color matching functions?
Go to the top of the page
 
+Quote Post
JRehling
post Sep 21 2006, 10:32 PM
Post #3


Senior Member
****

Group: Members
Posts: 2530
Joined: 20-April 05
Member No.: 321



QUOTE (slinted @ Sep 21 2006, 01:52 PM) *
If the VIMS data is in spectral power, then I would say that chromatic adaptation is still necessary since, as Don said, D65 isn't solar spectrum. It is what we see here on Earth.

If it is a reflectance dataset, then wouldn't that spectrum have to be multiplied by the spectrum of the D65 illuminant before being passed to the CIE color matching functions?


It's worse than that. wink.gif

In most of these cases, monitors are inherently dimmer than the sunlit surfaces we're talking about, and the relative sensory-system response to the three primary colors can and does shift as you increase/decrease illumination. So let's say some body out there simply reflected twice the green light it did blue light, reflecting those only in narrow bands, and it reflected no red at all. It would not necessarily capture the color of that object merely to have a display also emit twice the green light as blue; if the illumination is different, then a different ratio must be used to capture the ratio of color responses one would have seeing that object in daylight. An extreme example of this can cause the apparent relative brightness of two patches to depend on the illumination in which you view them -- even if the light sources both have solar spectrum! Another related phenomenon is that it is often possible to see sun dogs or dim rainbows if you are wearing sunglasses, only to have the color disappear when you take the sunglasses off.

I computed once that Uranus (and bodies beyond that, for the most part) has a luminance that a computer monitor can match, but everything closer to the Sun than that is brighter than your monitor could display. So achieving the colors one would perceive from a spaceship window requires some trickiness.
Go to the top of the page
 
+Quote Post
ugordan
post Sep 22 2006, 08:07 AM
Post #4


Senior Member
****

Group: Members
Posts: 3648
Joined: 1-October 05
From: Croatia
Member No.: 523



QUOTE (JRehling @ Sep 21 2006, 11:32 PM) *
I computed once that Uranus (and bodies beyond that, for the most part) has a luminance that a computer monitor can match, but everything closer to the Sun than that is brighter than your monitor could display.

Now that would be something... getting a totally dark room around you with an image of Neptune on the screen that is precisely as bright as the real thing and your eye is accustomed to the low light. I wonder how dim that'd appear. It would be difficult to calibrate, though. I imagine you'd need one of those Light Intensity Measuring Thingies (LIMTs).

There are CRT monitors nowadays that have a sort of "Magic Bright" function that kicks up the brightness substantially, I imagine accurate luminances could be reached even at Saturn, though I'm not so sure about Enceladus and Tethys, them being so high-albedo and all.


--------------------
Go to the top of the page
 
+Quote Post
ugordan
post Sep 22 2006, 06:41 PM
Post #5


Senior Member
****

Group: Members
Posts: 3648
Joined: 1-October 05
From: Croatia
Member No.: 523



I can understand the reasoning of not touching the illumination, not removing solar spectra from the cubes. The problem with that is solar spectrum turns out reddish and everything has a distinct red hue. Here are a couple of comparisons, left images show no spectrum messing, simple integration through the CIE XYZ functions.




I suppose if you stared long enough at scenes like this, the eye/brain would automatically adapt to the color and make a new white-point, making Enceladus (uppermost left image) appear white again. The problem is, this is not gonna happen on a computer screen.

You be the judge of what looks more reasonable.


--------------------
Go to the top of the page
 
+Quote Post
tedstryk
post Sep 22 2006, 09:12 PM
Post #6


Interplanetary Dumpster Diver
****

Group: Admin
Posts: 4404
Joined: 17-February 04
From: Powell, TN
Member No.: 33



For Jupiter and Saturn, I am biased toward the right hand images based on experiences with a telescope, but this brings in a whole new host of visual issues...


--------------------
Go to the top of the page
 
+Quote Post
ugordan
post Sep 25 2006, 08:02 PM
Post #7


Senior Member
****

Group: Members
Posts: 3648
Joined: 1-October 05
From: Croatia
Member No.: 523



I've been experimenting with cleaning up the VIMS visual channel a bit. I've found a cube that captures the night side of Titan with the entire 64x64 hires field of view that served as an augmented dark background to remove the pesky noise. It turned out better than I hoped. Unfortunately, the noise doesn't seem to be as static as I thought. Most of the noise is removed, but some isn't.

To illustrate just what mess the VIMS cubes are (including their calibration data), here are a couple of before and after images:



The bottom image is one of the highest resolution views of Iapetus by VIMS. The noise is disastrous. You can actually see it's mostly coincident with the ring image above.
My dark current removal model assumes noise is linear with exposure, this seems to work well. The background will need some touching up still.

I emailed one of the guys responsible for the PDS volumes, he said he'll look into it and get back to me. We shall see, though I'm not getting my hopes up.


--------------------
Go to the top of the page
 
+Quote Post

Posts in this topic
- ugordan   Processing VIMS cubes   Sep 10 2006, 07:51 PM
- - Malmer   its a really great "realitycheck" for th...   Sep 12 2006, 09:48 PM
|- - ugordan   QUOTE (Malmer @ Sep 12 2006, 10:48 PM) it...   Sep 13 2006, 08:56 AM
|- - Malmer   QUOTE (ugordan @ Sep 13 2006, 10:56 AM) I...   Sep 13 2006, 03:04 PM
- - dilo   Very nice work, gordan... now we see the colors...   Sep 12 2006, 10:21 PM
|- - ugordan   QUOTE (dilo @ Sep 12 2006, 11:21 PM) To m...   Sep 13 2006, 06:56 AM
|- - dilo   QUOTE (ugordan @ Sep 13 2006, 06:56 AM) T...   Sep 13 2006, 04:17 PM
- - CAP-Team   It looks as if the colors match Voyager's colo...   Sep 13 2006, 02:16 PM
- - Malmer   QUOTE (ugordan @ Sep 10 2006, 09:51 PM) I...   Sep 13 2006, 02:49 PM
- - DonPMitchell   How do you convert XYZ to RGB? Are you making sRG...   Sep 13 2006, 04:55 PM
|- - ugordan   Yes, I'm using that XYZ to RGB matrix. I had t...   Sep 13 2006, 08:33 PM
- - DonPMitchell   Very good. Gamma 2.2 is technically what should b...   Sep 14 2006, 04:07 AM
|- - ugordan   The pixels are linear as far as I know, a 12 bit A...   Sep 14 2006, 07:09 AM
|- - Malmer   QUOTE (DonPMitchell @ Sep 14 2006, 06:07 ...   Sep 14 2006, 09:30 AM
|- - Malmer   The pixels in the ISS images are not linearly save...   Sep 14 2006, 09:43 AM
||- - ugordan   I am very well aware of the ISS data specifics. Wh...   Sep 14 2006, 10:47 AM
||- - Malmer   QUOTE (ugordan @ Sep 14 2006, 12:47 PM) I...   Sep 15 2006, 12:50 PM
||- - ugordan   QUOTE (Malmer @ Sep 15 2006, 01:50 PM) On...   Sep 15 2006, 06:26 PM
||- - Malmer   it would be fun to paint spheres in the colors you...   Sep 15 2006, 11:40 PM
|- - DonPMitchell   QUOTE (Malmer @ Sep 14 2006, 02:30 AM) co...   Sep 14 2006, 03:39 PM
|- - Malmer   I think the strange piecewise funktion has its ben...   Sep 15 2006, 12:33 PM
|- - ugordan   QUOTE (Malmer @ Sep 15 2006, 01:33 PM) I ...   Sep 15 2006, 06:10 PM
- - DonPMitchell   One concept is to make an image that resembles wha...   Sep 16 2006, 01:56 AM
|- - Malmer   QUOTE (DonPMitchell @ Sep 16 2006, 03:56 ...   Sep 16 2006, 09:42 AM
|- - ugordan   A mosaic of Saturn's northern latitudes: App...   Sep 16 2006, 03:39 PM
||- - john_s   Just checked out your gallery, ugordan- your proce...   Sep 16 2006, 04:54 PM
||- - ugordan   QUOTE (john_s @ Sep 16 2006, 05:54 PM) Ju...   Sep 16 2006, 05:56 PM
|- - DonPMitchell   QUOTE (Malmer @ Sep 16 2006, 02:42 AM) so...   Sep 16 2006, 06:47 PM
|- - Malmer   QUOTE (DonPMitchell @ Sep 16 2006, 08:47 ...   Sep 16 2006, 09:02 PM
- - DonPMitchell   I also really like what you are doing. I hope you...   Sep 16 2006, 07:01 PM
|- - ugordan   QUOTE (DonPMitchell @ Sep 16 2006, 08:01 ...   Sep 16 2006, 09:37 PM
|- - DonPMitchell   QUOTE (ugordan @ Sep 16 2006, 02:37 PM) H...   Sep 17 2006, 04:17 AM
|- - ugordan   A couple of rough Saturn mosaics: The blob in the...   Sep 17 2006, 04:44 PM
|- - Malmer   I think it looks great. Very subtle beautiful colo...   Sep 17 2006, 07:17 PM
|- - ugordan   A few Jupiter images, fixed the greenish hue. It w...   Sep 19 2006, 05:21 PM
|- - Malmer   QUOTE (ugordan @ Sep 19 2006, 07:21 PM) A...   Sep 19 2006, 07:23 PM
|- - ugordan   The vertical stripes (most noticeable is a blue li...   Sep 19 2006, 07:37 PM
|- - Malmer   I guess you have to make an additional flatfield r...   Sep 19 2006, 08:16 PM
||- - ugordan   QUOTE (Malmer @ Sep 19 2006, 09:16 PM) I ...   Sep 20 2006, 07:24 AM
||- - Malmer   it was a typo. i actually ment darkcurrent. mixed ...   Sep 20 2006, 08:48 AM
||- - ugordan   Although the noise stays fixed, it changes intensi...   Sep 20 2006, 09:22 AM
||- - ugordan   Not particularly related to my processing, but her...   Sep 20 2006, 11:34 AM
||- - Malmer   QUOTE (ugordan @ Sep 20 2006, 11:22 AM) W...   Sep 20 2006, 03:23 PM
||- - ugordan   Here are a couple of... hm... questionable(?) resu...   Sep 20 2006, 07:38 PM
|- - tedstryk   Great work! As for the calibration data, it c...   Sep 19 2006, 08:18 PM
- - slinted   UGordon, first off, this is an amazing project you...   Sep 20 2006, 11:00 PM
|- - ugordan   Thanks, Slinted! I'll certainly have a loo...   Sep 21 2006, 07:46 AM
|- - JRehling   QUOTE (ugordan @ Sep 21 2006, 12:46 AM) I...   Sep 21 2006, 06:06 PM
|- - ugordan   JRehling, those are all valid, good points. I said...   Sep 21 2006, 07:16 PM
- - DonPMitchell   It is true that Judd and Vos made improvements to ...   Sep 21 2006, 12:55 PM
|- - ugordan   QUOTE (DonPMitchell @ Sep 21 2006, 01:55 ...   Sep 21 2006, 02:25 PM
- - DonPMitchell   Reflectance is independant of the illuminant. It ...   Sep 21 2006, 02:43 PM
- - DonPMitchell   There's a brown Moon for you, taken by Galil...   Sep 21 2006, 07:03 PM
- - slinted   Good points made by all, Part of my confusion on...   Sep 21 2006, 08:52 PM
|- - ugordan   VIMS cubes, their raw form, capture the brightness...   Sep 21 2006, 09:14 PM
|- - JRehling   QUOTE (slinted @ Sep 21 2006, 01:52 PM) I...   Sep 21 2006, 10:32 PM
|- - ugordan   QUOTE (JRehling @ Sep 21 2006, 11:32 PM) ...   Sep 22 2006, 08:07 AM
|- - ugordan   I can understand the reasoning of not touching the...   Sep 22 2006, 06:41 PM
|- - DonPMitchell   QUOTE (ugordan @ Sep 22 2006, 11:41 AM) Y...   Sep 22 2006, 07:06 PM
||- - ugordan   Be my guest. For starters, here's a dump of th...   Sep 22 2006, 07:45 PM
|- - tedstryk   For Jupiter and Saturn, I am biased toward the rig...   Sep 22 2006, 09:12 PM
|- - ugordan   I've been experimenting with cleaning up the V...   Sep 25 2006, 08:02 PM
|- - ugordan   Sure enough, I got a response from the VIMS team. ...   Sep 26 2006, 07:33 AM
- - edstrick   It probably boils down to too-much-to-do, not-enou...   Sep 26 2006, 09:49 AM
|- - ugordan   To reiterate what you said, on my asking if update...   Sep 26 2006, 02:27 PM
|- - ugordan   Here are a few quick results from the new PDS rele...   Oct 4 2006, 05:55 PM
- - Indian3000   First test to process VIMS cube ( gamma 2.2 ) v1...   Oct 11 2006, 11:53 AM
- - elakdawalla   Hey folks, the Rings Node has generated preview im...   Feb 18 2011, 10:20 PM
- - ngunn   A superb piece of work at TPS but we have to have ...   Apr 28 2011, 09:27 PM


Reply to this topicStart new topic

 



RSS Lo-Fi Version Time is now: 28th March 2024 - 08:24 PM
RULES AND GUIDELINES
Please read the Forum Rules and Guidelines before posting.

IMAGE COPYRIGHT
Images posted on UnmannedSpaceflight.com may be copyrighted. Do not reproduce without permission. Read here for further information on space images and copyright.

OPINIONS AND MODERATION
Opinions expressed on UnmannedSpaceflight.com are those of the individual posters and do not necessarily reflect the opinions of UnmannedSpaceflight.com or The Planetary Society. The all-volunteer UnmannedSpaceflight.com moderation team is wholly independent of The Planetary Society. The Planetary Society has no influence over decisions made by the UnmannedSpaceflight.com moderators.
SUPPORT THE FORUM
Unmannedspaceflight.com is funded by the Planetary Society. Please consider supporting our work and many other projects by donating to the Society or becoming a member.