Steve Albers textures are good,but raw

[danielj]

I don??t think that Steve Albers made a bad work,but textures like Rhea are using now are very raw,without any improvement or corrections of any kind in the image.I know that moon texture mapping is time demanding,but I don??t believe that only Fridger(a.k.a t00fri) is doing improved textures maps of saturnian moons.By the way,Titan and Iapetus textures are much better and realistic than Steve Albers??ones,but I know that t00fri is a busy scientist and don??t have much time free.Where are the other developers?
The next "test" will be Cassini??s Hyperion close flyby.I wonder if we could have a 2k texture map in a few weeks...

[hank]

Because of Hyperion's irregular shape, constructing a texture map from Cassini photos may be problematic.

- Hank

[danielj]

I know that,but this can be overcome because Phobos and Deimos are irregular too,and we have detailed 2k textures of both.And even from Voyager data,Celestia has a 1k texture map.Not mentioning Eros and Ida,that are both irregular,both we have 2k texture maps at least from Eros.

Because of Hyperion's irregular shape, constructing a texture map from Cassini photos may be problematic.

- Hank

[hank]

I know that,but this can be overcome because Phobos and Deimos are irregular too,and we have detailed 2k textures of both.And even from Voyager data,Celestia has a 1k texture map.Not mentioning Eros and Ida,that are both irregular,both we have 2k texture maps at least from Eros.

I'm not saying it can't be done; just that it won't be as simple or quick as with spherical worlds.

- Hank

[scalbers]

Greetings,

I thought I'd chime in some tidbits about recent textures - as I do feel honored to be mentioned in a topic title. I think we all have our strengths and weaknesses and it's good to compare results between Fridger, Cassini staff (e.g. Jason Perry), myself and hopefully others. At least for me, I have a certain amount of time to work on these and I'm prioritizing getting more images in vs quality of blending and keeping a realistic look true to the original photos and what a person would actually see if they were there.

I agree my Titan map has a lot of mismatch, though I can at least say I was the first one to add the recent flyby with the Lazy-H to a global map. This is one where I have been considering adding either Fridger's or Jason's map as a background to mine. One reason for hesitation is that the high frequency detail seems almost too strong due to enhancements that were done. Perhaps I've grown fond of that blurry raw look as that is what the spacecraft is actually seeing.

For Rhea, there are some big discontinuities though I do feel pride in getting a 4K map with good matching of the recent flyby images with each other. That side of Rhea looks really awesome now in Celestia. Recall the challenge of matching images with the sun overhead vs being right on the terminator.

For Iapetus, I have a different taste perhaps than Fridger. It's hard to describe, but I feel mine is more realistic looking. Fridger's may have fewer image blending discontinuities, yet to my eye it looks a bit too "artificial". I have kept more of the color variations from the original images in my map (albeit tricky with enhanced color originals) compared with Fridger.

I spent a lot of time recently on Enceladus and that perhaps is my best map at present. This is an intriguing satellite and a higher priority for me to do a good job on with my limited time.

With Dione, I'm just glad to get the images located correctly. It's tricky to map this satellite with large scale albedo variations to match along with the fine detail shadowing.

Anyway, hope that provides a glimpse of what I'm up to. Will be fun to see where we all head to next...

[scalbers]

Because of Hyperion's irregular shape, constructing a texture map from Cassini photos may be problematic.

- Hank

Phil Stooke is an expert on irregular satellite mapping with his Voyager based maps of satellites such as Mimas. He's around these days posting messages in the Mars forums of Unmannedspaceflight.com (formerly MER forum). Perhaps he'll take on Hyperion at some point?

I have started on Mimas, experimenting with considerations of a tri-axial ellipsoid. A truly irregular satellite like Hyperion is something I have yet to learn how to do.

[t00fri]

Greetings,
...
For Iapetus, I have a different taste perhaps than Fridger. It's hard to describe, but I feel mine is more realistic looking. Fridger's may have fewer image blending discontinuities, yet to my eye it looks a bit too "artificial". I have kept more of the color variations from the original images in my map (albeit tricky with enhanced color originals) compared with Fridger.
...

Steve,

perhaps it's worth emphasizing once more that I tried to keep my taste out of the game as much as possible by referring to NASA-published (natural, true, almost-natural,...) color imaging throughout. They have chosen to publish Iapetus in this 'chocolate'-brown color and I take it as it is, in my corresponding texture.

The clear advantage is that I can always cite NASA for it and don't have to refer to Schrempp's or Albers' subjective taste.

If they manage to improve that color display in the future, it's a snip of the finger and standard scientific practice to adapt to any improved published color imaging thereafter!

Bye Fridger

[scalbers]

Hi Fridger,

I'd like to clarify a couple of points about the color images for Iapetus. I actually have kept the chocolate brown color in my Iapetus texture. Of the four color images I've used, only one was a "near-true" color. I have (so far) kept my hue and saturation to match that one exactly. I have also used representative hue/saturation info from that image to colorize other grayscale images.

I have used two other published (here defined as from the JPL Photojournal) color images that were taken in enhanced color. I think we have both modified the color in these in a certain sense. I have decreased the saturation in a manner similar to what Bjorn Jonsson and I have done with Io. Perhaps it's partly semantics here, but if you had colorized these (treating as grayscale) it is also a change from what was originally published.

The one image I used that is perhaps more debatable is a raw RGB image combo that I used. Even this however is published in a certain sense by being posted (originally) as individual filter images on the JPL raw images website.

So I think it is partly a matter of "taste" or at least of opinion in coming up with the best strategy for utilizing these images. If we have differing ideas I think that is great to compare and neither one of us has an absolute say on the truth perhaps.

[t00fri]

Hi Fridger,

I'd like to clarify a couple of points about the color images for Iapetus. I actually have kept the chocolate brown color in my Iapetus texture. Of the four color images I've used, only one was a "near-true" color. I have (so far) kept my hue and saturation to match that one exactly. I have also used representative hue/saturation info from that image to colorize other grayscale images.

I have used two other published (here defined as from the JPL Photojournal) color images that were taken in enhanced color. I think we have both modified the color in these in a certain sense. I have decreased the saturation in a manner similar to what Bjorn Jonsson and I have done with Io. Perhaps it's partly semantics here, but if you had colorized these (treating as grayscale) it is also a change from what was originally published.

The one image I used that is perhaps more debatable is a raw RGB image combo that I used. Even this however is published in a certain sense by being posted (originally) as individual filter images on the JPL raw images website.

So I think it is partly a matter of "taste" or at least of opinion in coming up with the best strategy for utilizing these images. If we have differing ideas I think that is great to compare and neither one of us has an absolute say on the truth perhaps.

Steve,

I have only used the color mapping of the most recent Cassini Iapetus image. No earlier colorations!

My technique is to switch always the whole texture to grayscale, to match the shades of gray to the latest gray images and then to use the latest published (natural) color image for coloration of the whole texure.

At least my procedure is reproducable and only uses explicit color information from one credible template image.

Bye Fridger

[Dollan]

Steve...

I just wanted to say thank you for providing those images. I do try to keep up with your releases as Cassini reveals more and more of these remarkable satellites. I am primarily interested in simply displaying the current extent of our knowledge about these moons, and thus far your maps are the best that I've found. There are others, of course, who also do good work through their interpretations, but in the case of your work it is quite fun to see the moons slowly reveal themselves, bit by bit, over the months.

If you don't mind, there are a couple of Yahoo! Groups that I am a member of (one of them I own) where I'll post your web site. While not everyone there uses Celestia, most are quite interested in keeping up with the latest Saturnian discoveries.

Thanks again!

...John...

[t00fri]

Steve,

an obvious generalization of what I do concerning coloration would be this:

use n color images from the object of interest and generate a joint 1024 color palette from all of these. Use that palette for coloration.

In this method, perhaps some incorrect filter effects sort of average out...

Bye Fridger

[scalbers]

Steve...

I just wanted to say thank you for providing those images. I do try to keep up with your releases as Cassini reveals more and more of these remarkable satellites. I am primarily interested in simply displaying the current extent of our knowledge about these moons, and thus far your maps are the best that I've found. There are others, of course, who also do good work through their interpretations, but in the case of your work it is quite fun to see the moons slowly reveal themselves, bit by bit, over the months.

If you don't mind, there are a couple of Yahoo! Groups that I am a member of (one of them I own) where I'll post your web site. While not everyone there uses Celestia, most are quite interested in keeping up with the latest Saturnian discoveries.

Thanks again!

...John...

John,

I appreciate your interest - feel free to link to them in Yahoo, etc. Your publicity would be good to foster interest in Celestia, and also the project where I work that my image collection is used for called Science on a Sphere.

I enjoy sharing my incremental works in progress, so everyone can check out where I am. The understanding of course is that the various images are in differing states of completion. I bet there is a rich opportunity here to continue to polish these even for many years as Cassini sends more data and we discuss and improve in our processing techniques.

[scalbers]

Steve,

an obvious generalization of what I do concerning coloration would be this:

use n color images from the object of interest and generate a joint 1024 color palette from all of these. Use that palette for coloration.

In this method, perhaps some incorrect filter effects sort of average out...

Bye Fridger

Fridger,

Thanks for the explanation, I was wondering a bit more on how you might work with satellites where multiple "good quality" color images were available. I agree your method is fairly straightforward to explain. Perhaps my understanding of color palettes could be improved, though I wonder how limiting any assumptions are that imply a set value of hue/saturation to given values of intensity. I suppose I should try generating a map of hue/saturation for Iapetus and see how much they vary over uniformly bright or dark areas in the "near-true" color image.

I wonder how you might address this limitation (if it exists) in satellites such as Europa and Io. For Europa, I think there are some real hue variations (brown vs yellow) even in areas having the same brightness. Io of course is quite colorful and perhaps is more of a marked example of this.

Closer to "home", Rhea also poses this question to me. I've been weighing whether to use a higher resolution B&W image (and colorize it) or keep the present "natural-color" image I'm using preserving its variations in hue/saturation across the image (even in areas having the same brightness). Actually as I write this, there is another "win/win"
solution that occurs to me where one can use the low-res color info combined with the hi-res intensity info. I have done this to some extent in a couple of my maps. The Galileo people have done this quite a bit I believe in constructing some of the mosaics that I used.


Maybe Iapetus has more a constant brightness/color relationship than some of the other satellites do?

[scalbers]

Because of Hyperion's irregular shape, constructing a texture map from Cassini photos may be problematic.

- Hank

Phil Stooke is an expert on irregular satellite mapping with his Voyager based maps of satellites such as Mimas. He's around these days posting messages in the Mars forums of Unmannedspaceflight.com (formerly MER forum). Perhaps he'll take on Hyperion at some point?

To follow up on this I see Phil has posted today on Hyperion itself. Perhaps he's contemplating some mapping work at some point?

[danielj]

Where is Phill??s post?I can??t find it.

Because of Hyperion's irregular shape, constructing a texture map from Cassini photos may be problematic.

- Hank

Phil Stooke is an expert on irregular satellite mapping with his Voyager based maps of satellites such as Mimas. He's around these days posting messages in the Mars forums of Unmannedspaceflight.com (formerly MER forum). Perhaps he'll take on Hyperion at some point?

To follow up on this I see Phil has posted today on Hyperion itself. Perhaps he's contemplating some mapping work at some point?

[scalbers]

Where is Phill??s post?I can??t find it.

Try http://www.unmannedspaceflight.com/inde ... #entry8391