T00fri's Titan Flyby (03/31) Upgrade: DOWNLOAD NOW

[t00fri]

Hi all,

after the CYCLOPS team published yesterday a new hires
mosaic (centered at lat=1 deg North, long=21 deg West)
from the 03/31/05 Titan Flyby, I have started to implement
that spectacular new mapping information into my 4k Titan
texture.

Quite a bit of work for the weekend !

As you can see from the image below the new gray
composite overlay data still require a lot of smoothing and
leveling work. I already have performed the cylindrical map
projection, the high-quality rescaling and a careful alignment of the overlay.
Also I carefully cut out the new aligned individual images
from the lowres background in the CYCLOPS image
http://saturn.jpl.nasa.gov/multimedia/images/image-details.cfm?imageID=1469

Everything fits very precisely. The image below is reduced
by a factor of two relative to its natural size, and I
added about 50% transparency to the top (gray) layer for
better visibility:



The amount of visible detail is really great, as illustrated for example here:



Bye Fridger

[julesstoop]

Great work.
Nice to give us a peek in your kitchen (as they say in Dutch)

[t00fri]

Perhaps it is of some interest to potential image manipulators,
how accurately the alignment of the new hires layer from the
03/31 Titan flyby actually fits . Below I show a peripheral
portion of my original (orange background) and the
superimposed new hires layer (gray). For better visibility I
have adjusted the transparency of the gray overlay to 50%.
You may easily convince yourself that the continuity of
distinct hires features in the orange and gray layers
matches at the 1 pixel level! This is true, although in
the displayed peripheral area non-linear distortions due to
potential inaccuracies in the cylindrical projection are
amplified maximally!



Bye Fridger

[t00fri]

Hi all,

after a good deal of image manipulation work over the
weekend, my new upgraded Titan texture is essentially
finished. Everything is now seamless and smooth throughout the whole mosaic.

As obvious from my above posts, the new enhancements are major and
essentially the region around the "Lying H" is mapped in high
resolution during the 03/31 Titan flyby.

Before I offer the new texture for downloading, I want to do
a few final optimizations. Here is a glimpse already of what
the new stuff looks like.

Lots of intriguing new detail!




Enjoy,

Bye Fridger

[scalbers]

Fridger,

Looks like you're doing a great job with the Lazy-H. It's a bit tricky I think to combine the hi-res mosaics (having gaps) with the lower res ones. Will be interesting to see your released map. I think I have the link now to your Celestia Foundry site so I should be able to download it.

As you know I'm currently using your older Titan map as a background. Would it be OK if I do some experiments with your new map? I'd like to try either using it "as is" or as a background with any tinkering (or new images) I might think of.

I also made some general comments about mapping of Titan and other satellites in another thread today.

[t00fri]

Fridger,

Looks like you're doing a great job with the Lazy-H. It's a bit
tricky I think to combine the hi-res mosaics (having gaps)
with the lower res ones.

Not really, why?

I used /only/ the new CYCLOPS hires mosaic from 2 days
ago on the background of my last 4k texture that was based
also on a NASA publication. After doing the cylindrical
projection of the /whole/ mosaic, I cut the individual images
out of their lowres background /without/ disturbing their joint
alignment (see further above in this thread)!

The precision is excellent.

Will be interesting to see your released map.

Here is a partial view of the new texture around the "Lying
H". It is reduced by a factor of two. All transitions are very
smooth.



Next, a 850pixel mini version of my entire new 4k
texture. For better viewing I moved the central meridian to
the position you prefer.

I think I have the link now to your Celestia Foundry site so I
should be able to download it.

The TextureFoundry WEB-site is not upgraded. My new
textures you find always in obvious threads in 'users'.
You may use the search engine (looking for t00fri and titan).

As you know I'm currently using your older Titan map as a
background. Would it be OK if I do some experiments with
your new map? I'd like to try either using it "as is" or as a
background with any tinkering (or new images) I might think
of.

Anything you like, with proper citation, of course.

I also made some general comments about mapping of Titan
and other satellites in another thread today.

I answered already to some of it....

Bye Fridger

PS: Let me point out that I discovered an accidental incorrect
3 pixel vertical shift (out of 2048!) in my last 4k Titan
texture. This must have happened during some layer
manipulations. Correction of this improves somewhat the
agreement with the published Huygens landing position. The
new texture is of course corrected.

[maxim]

What do you call Lying-H? Is this the region around 'New Great Britain'?

maxim

[t00fri]

What do you call Lying-H? Is this the region around 'New Great Britain'?

maxim

Look at the cylindrical map: the black structure looks like the letter 'H' lying on its side...so that's what it's called.

Steve Albers calls it 'Lazy H'

Bye Fridger

[julesstoop]

What do you call Lying-H? Is this the region around 'New Great Britain'?

maxim

The inside does look somewhat like Great Britain indeed. I think you are both refering to the same region.

[rthorvald]

As you can see from the image below the new gray
composite overlay data still require a lot of smoothing and
leveling work. I already have performed the [b]cylindrical map
projection, the high-quality rescaling [...]

You know, showing a snapshot of the process - like this mosaic - really points out what a complex juggling act it is to produce this. I appreciate it very much, and the screenshots look fascinating.

Really hope i´ll have an opportunity to download it soon...

-rthorvald

[t00fri]

As you can see from the image below the new gray
composite overlay data still require a lot of smoothing and
leveling work. I already have performed the [b]cylindrical map
projection, the high-quality rescaling [...]

You know, showing a snapshot of the process - like this mosaic - really points out what a complex juggling act it is to produce this. I appreciate it very much, and the screenshots look fascinating.

Really hope i´ll have an opportunity to download it soon...

-rthorvald

Many thanks Runar...it's indeed a lot of work, but --as you
know very well-- its also a lot of fun to do

How do people feel about adding a bump-map file and a spec
file to my forthcoming update package?? Purists (like myself
) can always throw them away or comment them out.

Titan just looks better with a (moderate) bumpmapping and a
bit of "spec". I am aware, of course, that it's a little less
scientifically rigorous

Bye Fridger

[TERRIER]

How do people feel about adding a bump-map file and a spec
file to my forthcoming update package?? Purists (like myself
) can always throw them away or comment them out.

Titan just looks better with a (moderate) bumpmapping and a
bit of "spec". I am aware, of course, that it's a little less
scientifically rigorous

Bye Fridger

By all means do so.
I shall install them as alternate surfaces like I did with your last version.

[jestr]

Fridger,how do you make the bump maps,from the luminosity of each pixel?I have found a way to build a model of a planet/moon using the bump map as a displacement map,(this only works well on spherical objects though),but when I tried it with Iapetus,of course the Voyager mapped side looked rather high and smooth .I guess we'll have to wait to 2007 when Cassini goes past.Keep up the good work,Jestr

[t00fri]

Fridger,how do you make the bump maps,from the
luminosity of each pixel?I have found a way to build a model of a
planet/moon using the bump map as a displacement map,(this only
works well on spherical objects though),but when I tried it with
Iapetus,of course the Voyager mapped side looked rather high and
smooth .I guess we'll have to wait to 2007 when Cassini goes
past.Keep up the good work,Jestr

Jestr,

well, without a level map, anything that can be done is clearly a
compromise...

I usually start from the luminosity map but then go through rather
elaborate steps to make things as smooth /and/ noiseless as possible.
Typically I produce various copies of the luminosity map, apply
different sized Gaussian blurs to each one and then combine them via
various modes like 'overlay' or 'multiply'. After each step, the resulting
luminosity has to be readjusted, of course. This procedure is known to
be an effective noise-killer and tends to give very smooth gradients.


Then I may locally work on features that I want to be specifically
elevated or depleted through fuzzy selection, possible inversion,
brightness and contrast adjustments.

A typical result from such a procedure is illustrated in this image of my
Iapetus texture. Here, extreme smoothness was essential to model the
smooth white ice flows over the brown crater landscape in
agreement with the respective NASA photo. Here is a reminder of how
my result looked like...



Bye Fridger

[PlutonianEmpire]

Umm, do you think we'll ever be able to map Titan's north pole before the Cassini mission ends?