Printable Version of Topic

Click here to view this topic in its original format

Unmanned Spaceflight.com _ Mars Odyssey _ new THEMIS color releases

Posted by: DDAVIS Jun 22 2006, 02:23 PM

First of a new series:

http://themis.asu.edu/zoom-20060622a

Posted by: djellison Jun 22 2006, 02:30 PM

Bloody hell Don, I don't know how you get such smooth data out of those released files!

Posted by: Stu Jun 22 2006, 04:47 PM

ohmy.gif ohmy.gif ohmy.gif

Amazing work Don, thanks... THEMIS doesn't get NEARLY as much credit or exposure as it deserves, methinks...

Posted by: climber Jun 22 2006, 06:42 PM

If I read it correctly, these are true colours. Details and the whole picture are just magnificent. True artistic work.

Posted by: DDAVIS Jun 22 2006, 07:51 PM

QUOTE (climber @ Jun 22 2006, 06:42 PM) *
If I read it correctly, these are true colours. Details and the whole picture are just magnificent. True artistic work.



Thanks! I have actually done less 'editorial' colour adjustment in this series than with the Meles Chasma image, the goal being to try to show what the camera was seeing while removing the distracting light leak stripes. When the judgement has to be made on supressing bright or dark discontinuities in an image, the direction this would bring the color balance toward or away from the known range of Martian colors recieves consideration. There are some images where the 'purple' cast common to RGV images shows in dark parts of the image, and others where the scene may have a haze overlay giving a blueish to violet 'wash' over much of the scene, but not all. Anything that looks like it's not an obvious camera related artifact is left in. Seeing subtle color variations within a frame are among the strengths of the THEMIS camera, and images showing such are paramount among the ones I look for.

A slightly dated page on color filters and color images can be seen here:
http://www.donaldedavis.com/2003NEW/CLRWORK/CLRTEST.html

The examples suggest something of the color fidelity of various spacecraft cameras, including the R(green faked from clear)B New Horizons imaging system.

Don

Posted by: edstrick Jun 23 2006, 10:43 AM

The sad fact is that the Mars Odyssey cameras have fairly serious design defects that were not caught by enough testing and review, possibly another victim of smaller-faster-better-cheaper. We've had a consistant problem over the years designing cameras that can do more than just good monochrome imagery of low-color-contrast planetary targets.

Voyager, Galilleo and Cassini have done a pretty good job (earlier missions were on the slope of the learning curve), but missions like NEAR (dinky little, noisy, non-square-pixel-CCD), Global Surveyor (wide angle camera's red/blue color data is hard/impossible to align perfectly and the blue channel's noisy), Odyssey (multispectral data's got terrible uncalibratable (more or less) internal reflection defects), Stardust (outgassing on optics, stuck filter wheel), Deep Impact (grossly out-of-focus hirez camera)... just don't perform up there where they should have.

<sigh... rant.. rave..>

Posted by: djellison Jun 23 2006, 11:42 AM

HRSC's all right wink.gif

Doug

Posted by: Nix Jun 24 2006, 10:49 AM

ohmy.gif Amazing work Don!

Nico

Posted by: AlexBlackwell Jun 27 2006, 05:33 PM

QUOTE (edstrick @ Jun 23 2006, 10:43 AM) *
The sad fact is that the Mars Odyssey cameras have fairly serious design defects that were not caught by enough testing and review, possibly another victim of smaller-faster-better-cheaper. We've had a consistant problem over the years designing cameras that can do more than just good monochrome imagery of low-color-contrast planetary targets.

Voyager, Galilleo and Cassini have done a pretty good job (earlier missions were on the slope of the learning curve), but missions like...Odyssey (multispectral data's got terrible uncalibratable (more or less) internal reflection defects...

I'm not sure if it's been mentioned previously but McConnochie et al. have a paper entitled "Calibration and in-flight performance of the Mars Odyssey Thermal Emission Imaging System visible imaging subsystem (THEMIS VIS)" (http://marswatch.tn.cornell.edu/THEMIS-VIS-cal_submitted.pdf) that, as I understand it, will be published online tomorrow in JGR-Planets.

Posted by: djellison Jun 27 2006, 07:19 PM

A similar design was used for MARCI as well, sort of half-push-broom-half-discreet-ccd, and looking through that paper, it looks like at utter barsteward to process - why would one choose that over a normal pushbroom?

Doug

Posted by: mcaplinger Jun 27 2006, 07:56 PM

QUOTE (djellison @ Jun 27 2006, 12:19 PM) *
A similar design was used for MARCI as well, sort of half-push-broom-half-discreet-ccd, and looking through that paper, it looks like at utter barsteward to process - why would one choose that over a normal pushbroom?

We call this a "pushframe" system. An N-color pushbroom system needs N separate linear CCDs and N different signal chains. An N-color pushframe needs one area CCD and one signal chain. MARCI weighs 350 grams; it would be simply impossible to build a linear pushbroom system for anything like that mass.

As for the difficulty of processing: who cares? Computers are fast and cheap.

See http://www.msss.com/mro/marci/images/2006/04/13/marci_detail/index.html

Posted by: djellison Jun 27 2006, 08:45 PM

You see - this is why we need you around here smile.gif

I must admit - the MARCI Orbit 7/8 imagery did make me go "ooo - that's all a bit complicated...but if it works..what the hell"

Doug

Posted by: AlexBlackwell Jun 28 2006, 09:14 PM

QUOTE (AlexBlackwell @ Jun 27 2006, 05:33 PM) *
I'm not sure if it's been mentioned previously but McConnochie et al. have a paper entitled "Calibration and in-flight performance of the Mars Odyssey Thermal Emission Imaging System visible imaging subsystem (THEMIS VIS)" (http://marswatch.tn.cornell.edu/THEMIS-VIS-cal_submitted.pdf) that, as I understand it, will be published online tomorrow in JGR-Planets.

As http://www.agu.org/pubs/crossref/2006/2005JE002568.shtml.

Posted by: edstrick Jun 29 2006, 04:31 AM

A FUNDAMENTAL problem pushing Mars camera designers toward pushframe systems is the relatively low light levels and low to very low contrast levels of surface imaging at moderate to low sun angles through that hazy atmosphere. The dwell-time of a single-pixel-wide, gadzillion-pixels-long detector's pixels on the surface is "NanoSized" as the resolution of the camera gets "SuperSized".

A pushframe system, perfectly (to a subpixel scale) perpendicular to the moving field-of-view, multiplies that dwell time by the number of pixels the detector is wide. Signal-to-noise ratio goes from c__p to good or realliy fine. That's particularly important for color work on a low-color-contrast (in many areas) planet.

Whatever lead to the multiple internal reflection problems and whatever else is wrong or marginal in the Themis camera, seems to be fixed or avoided in ouir spiffy new "Ancient-Martian-Astronauts-Espionage-Orbiter" ... oops... I mean Mars Reconnaisance Orbiter's supercam. The test images taken under unfavorable conditions show vanishingly small brightness or color artifacts in the released data (made using preflight or first-draft inflight calibration files and software). You can see some stitching-errors in the overlaps between adjacent CCD's in the multi-detector system, but that's not bad at all for first-try tests.

Posted by: mcaplinger Jun 29 2006, 06:37 AM

QUOTE (edstrick @ Jun 28 2006, 09:31 PM) *
A pushframe system, perfectly (to a subpixel scale) perpendicular to the moving field-of-view, multiplies that dwell time by the number of pixels the detector is wide.

I think you're confused. What you're describing is called Time Delay Integration or TDI, and has nothing to do with pushframe, which is a multispectral technique. It's true that these can be combined, but I don't know of that being done in any previous flight instrument (though our JunoCam instrument will do it to a limited degree.)

Posted by: edstrick Jun 29 2006, 08:13 AM

I stand corrected. My understanding remains that the HiRes cam does cross-CCD integration, but I'd have to recheck the specs.

Posted by: djellison Jun 29 2006, 08:58 AM

Yes - HiRISE does TDI - (128 lines I think)

Essentially as eash row 'dwells' on the target, it adds the photons, then passes the count onto the next row of the 128 rows....thus increasing SNR by, in theory, 128 times. Of course there is noise and some losses within the process - but that's the theory as I understand it.

MOC NA, MOC WA, HRSC and CTX are, as I understand it, single line Pushbrooms. HiRISE a Pushbroom-with-TDI, and MARCI / THEMIS a Pushframe design. Discreet framing cameras were on board the Vikings, Mariners, and the crap SRC channel on HRSC smile.gif

Doug

Posted by: edstrick Jun 29 2006, 09:23 AM

The absolutely critical thing with a camera like HiRISE is the image must smear nearly perfectly across the detector parallel to the cross-frame rows of pixels and the camera must step the "bucket" of electrons along the column in near perfect time with the image motion, and nothing can jiggle during the imaging. "Nearly perfectly" is probably somewhat better than 1/4 pixel or so. That would be measurable but not obviously degrade the image.

That HiRISE did so well from a grossly non-nominal elliptical orbit under low to extremely low sun illumination angles and did so with "first try" image aquisition software and decalibration files and software says that barring hardware failure, we've got ourselves a winner.

The only sad thing is that the "phace-phreaks" will still be able to "discover" untold signs of archaeological wonders at and well beyond the resolution limits of the images, no matter how high the resolution is. Hoaxland would be able to discover signs of artificial intelligence in the artifacts of an image of a single electron, I'd swear!

Powered by Invision Power Board (http://www.invisionboard.com)
© Invision Power Services (http://www.invisionpower.com)