It is currently Tue, 23-05-17, 7:02 GMT

All times are UTC




Post new topic Reply to topic  [ 6 posts ] 
Author Message
 Post subject: LRO WAC color map
PostPosted: Sun, 15-11-15, 3:01 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 2:32 GMT
Posts: 425
Location: South Korea
At KARI, among other things I'm working on building an ISIS 3-based lunar image processing pipeline. Out of the blue, I came across this rather interesting and possibly heavily-overlooked article from 2014 about creating a Hapke photometric parameter map of nearly the entire Moon, a world first!

http://lroc.sese.asu.edu/posts/801

The result is not only semi-global Hapke parameter maps enabling photometric correction at local scales, but also nice full-color maps from -55s to 55n latitudes (the polar regions present a problem because of permanently-shadowed regions) that register perfectly with WAC and LOLA topographic data (i.e., normal maps). Here is what the color map looks like, and the previous WAC monochrome map that we've all been using, for comparison:

Attachment:
WAC_color.jpg
WAC_color.jpg [ 303.46 KiB | Viewed 1206 times ]
Attachment:
WAC_mono.jpg
WAC_mono.jpg [ 302.68 KiB | Viewed 1206 times ]


Here are the Hapke parameter maps from the paper, which also contains a ton of interesting discussion on lunar photometry and geology:
Attachment:
hapke_maps.jpg
hapke_maps.jpg [ 118.15 KiB | Viewed 1206 times ]


Not long ago, Fridger showed us how spectacular color maps can be with his adaptation of Chang'e 2 imagery.
It's clear that we need a good color map that registers well with topographic data. As you can see, crater rays are much more visible in the WAC color map and the different tones of color within maria are visible.

Here is an interactive preview:

http://bit.ly/1Mez5YW


Top
 Profile  
 
 Post subject: Re: LRO WAC color map
PostPosted: Sun, 15-11-15, 9:45 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4442
Location: Hamburg, Germany
Hi Dawoon,

thanks for reporting this very interesting discussion about the effects of global Hapke parameter maps and for the reference to the original papers involved. I doubt that many readers know what Bruce Hapke -- a LROC team member-- did in 2012 ;-) .

Looking at the distinct color differences in the adjacent mares of your above example rendering from the interactive preview, I would rather guess that the Hapke photometric model furnishes sort of an enhanced color representation rather than natural (<-> visual) color. I did not find any statements about this crucial issue in the paper.

However --as a keen amateur astronomer -- I have observed the moon and its visual colors over many many years with a variety of excellent telescopes... I have never seen such marked and abrupt color changes from dark blue to dark brown in adjacent mares.

[Please, click on image for a clearer color view]
Attachment:
Hapke.jpg
Hapke.jpg [ 68.2 KiB | Viewed 1193 times ]


As you know very well, in Celestia and currently also in celestia.Sci we always try to render in Natural color to match the vision of the human eye.

For celestia.Sci this may change at a later stage, once color profiles and the transformations among them have been completely implemented. Then Hapke photometry could become a very important tool, at least for the imaging of the Moon...

Cheers,
Fridger

_________________
Image


Top
 Profile  
 
 Post subject: Re: LRO WAC color map
PostPosted: Mon, 16-11-15, 5:27 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 2:32 GMT
Posts: 425
Location: South Korea
I think that the WAC color map produced by photometric correction is not in fact "enhanced" and does represent a physical model. You are correct however in that it is not "natural" either, in the sense that the color map is not what we see. But this applies to any color texture -- for example the Chang'e 2 texture is not natural on its own and only looks natural once we apply shading using a photometric function.

The colors in the WAC map I think are supposed to represent what we might see if all of the special lunar photometric characteristics such as coherent backscatter have been normalized, resulting in "pure" albedos being represented.
To obtain a natural look, one would have to apply the Hapke photometric map to the WAC color map when rendering. Currently a single Hapke function is applied globally to the entire Moon; this would have to change so that the fragment shader reads the WAC color map values and Hapke parameter maps to compute the lighting function correctly at each fragment.


Top
 Profile  
 
 Post subject: Re: LRO WAC color map
PostPosted: Mon, 16-11-15, 9:11 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4442
Location: Hamburg, Germany
dirkpitt wrote:
I think that the WAC color map produced by photometric correction is not in fact "enhanced" and does represent a physical model. You are correct however in that it is not "natural" either, in the sense that the color map is not what we see.


Did you come across any quotation in the related literature, where it is stated that the Hapke model is closer to what NASA calls Natural color than to what NASA calls Enhanced color?

The coloration technique I used for the Chang'e 2 texture and actually for many of my textures employs a reference map into Natural color. The method works usually quite satisfactorily. What I can say is that e.g. my colored Chang'e 2 texture agrees pretty well visually with my colored photos I made of the Moon with telescopes.

You may also compare how this GIMP coloration method generates a Natural color approximation to Pluto. The result (see my respective textures) is certainly much closer to Natural color than to Enhanced color that is also available from NewHorizon imagery.

Fridger

_________________
Image


Top
 Profile  
 
 Post subject: Re: LRO WAC color map
PostPosted: Fri, 20-11-15, 0:09 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 2:32 GMT
Posts: 425
Location: South Korea
I agree with you Fridger in that the colors do look somewhat non-natural in the WAC color product with Hapke normalization, so I did some more digging on my own to find out why.
I can't find any literature that comments on whether the product is intended to be natural or not, but I have found a couple of related things that may shed some light:

The WAC color product with Hapke normalization is not the only WAC color map out there. Here is another one from 2011 (https://lpod.wikispaces.com/October+16,+2011):
Image

That image, while not very clean, also shows the same strange color contrast in maria.

But note the wavelengths used to represent rgb for the images:
LPOD 2011: R 566nm, G 360nm, B 321 nm
Hapke: red:689 nm, green:415 nm, blue:321 nm

However, according to CIE 1931 the rgb primaries we should be seeing on monitors are: 650, 530, 460 nm (http://people.virginia.edu/~smb3u/Color ... ode17.html).

So I think that the combination of the facts that 1) the WAC color maps are being displayed before they are lit and shaded, and 2) use of different RGB primaries than CIE 1931, are causing the WAC color maps to look unnatural.
FYI the color bands of the LROC WAC are: 415, 566, 604, 643, 689 nm. So one could try to select say, bands 643, 566, 415 instead to more closely approximate CIE rgb primaries.


Top
 Profile  
 
 Post subject: Re: LRO WAC color map
PostPosted: Fri, 20-11-15, 10:01 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4442
Location: Hamburg, Germany
It is quite usual for spacecraft cameras / imaging spectroscopes to include ultra-violet and | or infra-red (filter) components in their imaging. See e.g. also NH cameras (Ralph, ALICE) or the cameras on board of the Cassini spacecraft. A good aproximation to Natural color is then obtained by transformation from these more-or-less shifted color axes. As a reminder, in case of Titan, IR imaging was required to penetrate the thick atmospheric haze.

The LROC camera filters are here: http://lroc.sese.asu.edu/about/specs. What I used for Chang'e 2 data was UVVIS 5 band color basis from the Clementine mission.

_________________
Image


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 6 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group