IPB

Welcome Guest ( Log In | Register )

38 Pages V   1 2 3 > »   
Reply to this topicStart new topic
Reprocessing Historical Images, Looking for REALLY big challenges?
GregM
post Apr 21 2005, 11:26 PM
Post #1


Member
***

Group: Members
Posts: 123
Joined: 21-February 05
Member No.: 175



.
Go to the top of the page
 
+Quote Post
Gsnorgathon
post Apr 22 2005, 08:05 AM
Post #2


Member
***

Group: Members
Posts: 257
Joined: 23-January 05
From: Seattle, WA
Member No.: 156



I have to admit I'm a real sucker for those "first images of..." They're a great reminder of how far we've come.

Ted Stryck has a whole slew of cool stuff.

Don P. Mitchell has some nice reprocessed Venera images.
Go to the top of the page
 
+Quote Post
OWW
post Apr 22 2005, 08:16 AM
Post #3


Member
***

Group: Members
Posts: 708
Joined: 28-September 04
Member No.: 99



Cool Mariner 4/6/7 ( and more ) stuff:
http://members.tripod.com/petermasek/marinerall.html

Some additinal Mariner 6/7 material:
http://cps.earth.northwestern.edu/index.html
Go to the top of the page
 
+Quote Post
4th rock from th...
post Apr 22 2005, 12:14 PM
Post #4


Member
***

Group: Members
Posts: 378
Joined: 21-April 05
From: Portugal
Member No.: 347



The problem is having the original datasets for the Pioneers and Veneras.
Otherwise one can only work from published JPGs ou images scanned from paper...


--------------------
_______________________
www.astrosurf.com/nunes
Go to the top of the page
 
+Quote Post
GregM
post Apr 22 2005, 04:01 PM
Post #5


Member
***

Group: Members
Posts: 123
Joined: 21-February 05
Member No.: 175



.
Go to the top of the page
 
+Quote Post
tedstryk
post Apr 22 2005, 04:24 PM
Post #6


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4393
Joined: 17-February 04
From: Powell, TN
Member No.: 33



Don Mitchell has gotten ahold of the original Venera surface transmissions, and is looking for the Venera 9/10 orbiter image data. His work is based on the original 9-bit imagery, which really helps the quality.
http://www.mentallandscape.com/C_CatalogVenus.htm
http://www.mentallandscape.com/C_Catalog.htm
As for the Pioneers, the best thing would be to go back to the original transmissions from the photopolarimeter scans and reconvert them into images using modern software. However, they are 6-bit, so the quality will be limited. But I imagine modern processing could do much more to fix stepping errors and the effects of spacecraft motioin - it often took 40+ minutes to scan a full image of Jupiter and Saturn!


--------------------
Go to the top of the page
 
+Quote Post
gpurcell
post Apr 22 2005, 06:55 PM
Post #7


Member
***

Group: Members
Posts: 242
Joined: 21-December 04
Member No.: 127



QUOTE (GregM @ Apr 22 2005, 04:01 PM)
QUOTE (4th rock from the sun @ Apr 22 2005, 12:14 PM)
The problem is having the original datasets for the Pioneers and Veneras.
Otherwise one can only work from published JPGs ou images scanned from paper...
*



Yup, that's the key - the original datasets. Trying to enhance already poor existing images is kinda like trying to make a silk purse out of a sow's ear. Well, kinda, I'm sure something can be done with them to improve on them, but getting the original datasets is the proper way.

JPL, Ames, Goddard, LPL, et. al. isn't gonna do it. I was thinking of seeing some sort of relationship being set up where such legacy datasets might be passed on to expert amateurs or students working on a thesis, or some other sort of thing.

Just a hope.
*



Greg, shouldn't the original datasets be public property? Could a FOI request be filed for them?
Go to the top of the page
 
+Quote Post
tedstryk
post Apr 22 2005, 08:02 PM
Post #8


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4393
Joined: 17-February 04
From: Powell, TN
Member No.: 33



In the case of the Venera data, that is a Russian thing, so it is a little trickier. The raw tapes were public domain, but again, since foreign, wouldn't be covered by FOI. Pioneer is another story. I think the problem in that case lies that, if the original transmissions even survive, they may be in an archaic format on brittle old magnetic tapes that need special equiptment to be copied, if that is even possible. Also, for Mariner IV, much of the original digital data is missing.


--------------------
Go to the top of the page
 
+Quote Post
deglr6328
post Apr 22 2005, 11:56 PM
Post #9


Member
***

Group: Members
Posts: 356
Joined: 12-March 05
Member No.: 190



QUOTE (tedstryk @ Apr 22 2005, 08:02 PM)
In the case of the Venera data, that is a Russian thing, so it is a little trickier.  The raw tapes were public domain, but again, since foreign, wouldn't be covered by FOI.  Pioneer is another story.  I think the problem in that case lies that, if the original transmissions even survive, they may be in an archaic format on brittle old magnetic tapes that need special equiptment to be copied, if that is even possible.  Also, for Mariner IV, much of the original digital data is missing.
*


The US data shouldn't be TOO impossible to get. check here? smile.gif
Go to the top of the page
 
+Quote Post
tedstryk
post Apr 23 2005, 03:04 AM
Post #10


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4393
Joined: 17-February 04
From: Powell, TN
Member No.: 33



I have not come across original Pioneer image data on the PDS. The NSSDC holds prints of all Pioneer images (made using the original processing of the data), as well as polarization data and copies of the images on microfiche. But I have no idea where the original transmissions are, if they are still around.


--------------------
Go to the top of the page
 
+Quote Post
GregM
post Apr 23 2005, 04:28 AM
Post #11


Member
***

Group: Members
Posts: 123
Joined: 21-February 05
Member No.: 175



.
Go to the top of the page
 
+Quote Post
edstrick
post Apr 23 2005, 09:25 AM
Post #12


Senior Member
****

Group: Members
Posts: 1870
Joined: 20-February 05
Member No.: 174



The digital data for the final versions of <most?> of the Mariner 4 mars images are available in many university engineering or aerospace department libraries. *** IN PAPER COPY! *** The Mariner Mars 1964 Project Report <in multiple volumes> contained a volume entitled something like "picture element matricies", which contained PRINTOUTS of the 200 lines and 200 samples of 6 bit data for each processed image.

I've wanted for years to scan and OCR the data and make digital versions of the images, but ..... time and effort .....

Note that the mariner 4 images were taken in alternating red and green filters, with every third image not recorded. So the pictures are: green, red, missing, red, green, missing, green, red, missing etc. The corner of one image of each red/green pair OVERLAPPED by about 100 pixels, so there is some color data from the Mariner 4 images.

All that was done for the final project report was not entirely accurate registration of the overlapping image segments and either differencing or ratioing of the images to separate color differences from brightness differences. Nothing was published in any form as a color picture from that data to my knowledge.
Go to the top of the page
 
+Quote Post
edstrick
post Apr 23 2005, 09:50 AM
Post #13


Senior Member
****

Group: Members
Posts: 1870
Joined: 20-February 05
Member No.: 174



Mariner 6 and 7 returned two "primary" image datasets of Mars. Far Encounter and Near Encounter. Far encounter data was narrow angle only, near encounter was alternating wide and narrow angle images. The narrow angle camera was monochrome, with <I think> a yellow filter to "cut haze", while the wide angle camera had a 4 position filter wheel. I think red/green/blue/green.

The spacecraft had dual tape recorders, derived from the Mariner 4 hardware. Unfortunately, it did *NOT* have the capability to recorde the entire dataset in digital form.

All data was recorded on an *ANALOG* tape at full resolution, after being sharpened with a circuit that enhanced fine details. Near encounter analog data was recorded after being passed through an automatic gain control filter <which totally trashed absolute intensity values for the data>. Unfortunately, the taped data was degraded by significant in-flight buildup of tape oxide on the playback <record/playback?> heads, resulting in horizontal streaky noise and banding and occassional analog dropouts in the images.

Every seventh pixel was recorded on the DIGITAL tape recorder without automatic gain control, after having the two most significant bits chopped off, which badly "contoured" the digital data. To make matters worse, the middle 1/4 or 1/5 of each image's digital data was not recorded, spectrometer data being transmitted during that gap <and between the end and beginning of each scan line as well>. BUT..... every 4'th digital pixel in the center gap was trasmitted in real time!....

The efforts to reconstruct the images were described in the Mariner 69 special report issue of the Journal of Geophysical Research, published in 1970 or early 71. They were largely successful, producing 2 versions. One was from the analog data, and was a "Maximum Discriminability" version of the data, best showing fine details. The other was radiometrically decalibrated using preflight and some inflight data on camera background and shading and spectral response, starting from the reconstructed final images. These show the real contrasts of the martian surface better, and the regional shading of the albedo markings.

The wide angle images were taken in overlapping swaths through alternating color filters, and some color information is clearly present in the images. The viewing geometry changed from frame to frame as the field of view was something like 15 degrees. I have never seen attempts to generate color images using the overlapping wide angle close encounter images.

Mariner 69 also returned a "secondary" image dataset... the every 7'th pixel data <missing that central stripe> was transmitted in real time. Most of the far encounter data includes at most slivers of the planet's limb, as the disk was within the central stripe, but Mariner 7 transmitted a unique set of WIDE ANGLE "Late Far Encounter" data with the filter wheel stepping. This data was not recorded for playback. There is a picture floating around made from this data of a half mars <cut off by that central stripe> of the Sinus Meridiani area, in color, with mabye twice or so the resolution of most any earthbased mars imagery from that period.
Go to the top of the page
 
+Quote Post
edstrick
post Apr 23 2005, 10:14 AM
Post #14


Senior Member
****

Group: Members
Posts: 1870
Joined: 20-February 05
Member No.: 174



Mariner Mars 1971 was to be a two spacecraft mission. One spacecraft was to be placed in a high-inclination orbit optimized for mapping, the other in an orbit optimized for "variable features" and atmosphere studies, with viewing geometry repeating closely evey few orbits. Mariner 8 went in the drink due to launch failure, and the Mariner 9 mission was redisigned to do both tasks, with the emphasis on mapping.

The imaging system was an upgrade of the Mariner 69 system, but with a new, all digital, tape recorder, capable of storing some 33'ish 800x600 or so pixel images. As before, the narrow angle camera was monochrome, with I think a yellow "haze filter". The wide angle camera had a filter wheel with a quite nice set of (I think 8) filters: Violet/Blue/Green/Red, 3 orange polarizing filters, and maybe Clear.

When Mariner 9 got to Mars in Nov 71, the greatest global dust storm *EVER* observed on the planet, which had started in Aug or Sep 71 was still essentially obscuring the entire planet. The dust was starting to settle, but... the only things visible with any contrast were 4 dark spots (the tharsis volcanos, sticking out of the dust) and the small summer south polar cap.

Mariner started it's pre-programmed mapping sequences, but it was immediately obvious that they were 3/4 useless, and a couple of makeshift interim observation sequences were developed and put in operation. A better dust storm and south polar observation sequence followed about a month after arrival, and included test mapping pics to check on atmosphere clearing. Things progressively improved and mapping was started in <I think> early January 72.

*UNFORTUNATELY*.... by then, due to a hardware failure, the filter wheel on the wide angle camera had stuck on one of the orange polarizing filter positions. The data was acceptible for mapping purposes, but any further color imaging was lost. Some color images of dark albedo markings at high southern lattitudes were generated as well as color pictures of the south polar cap and surrounding hazy atmosphere and dimly visible surface, but that's all I've seen.

Mariner 9's camera had one severe design limitation that made quantitative decalibration of the images very difficult <The Mariner 69 cameras had the same problem>. Each vidicon image exposure was erased with repeat scans of an electron beam after electron beam readout from the camera, but the erasure was incomplete. Ghosts of previous images with some 5 to 10% of the brighness of the previous image, then fading slowly, persisted in subsequent images. In mapping strips of images, that wasn't terrible, but when the ghost images contained the bright limb of the planet, they were awful. The strength of each ghost depended on the pixel location, and the brightness of the previous image and the brightness of the current image. The resulting images were significantly degraded cosmetically and scientifically by the residual images.

Mariner Venus/Mercury 1973 (Mariner 10) solved the problem with perfect brute force engineering. Each of the twin cameras contained a ring of several small "wheat bulb" light bulbs! After each exposure was read out, the light bulbs INSIDE the camera were turned on, totally and uniformally saturating the vidicon surface with light. The cameras then erased the vidicon repeatedly, before taking a new exposure. Residual image was reduced to essentially undetectible levels.

The Viking Orbiters and Voyagers did the same thing. The "Light Flood" could be turned off when desired, as it left a background level of "glow" noise in the images, and when Voyagers, in particular, were doing long exposure low-light-level imaging, the light flood was "off".
Go to the top of the page
 
+Quote Post
tedstryk
post Apr 23 2005, 12:24 PM
Post #15


Interplanetary Dumpster Diver
****

Group: Moderator
Posts: 4393
Joined: 17-February 04
From: Powell, TN
Member No.: 33



The reason so much of Mariner 6/7 imagery was sent back in a lossy way or in analogue was the extremely limited storage capacity. Mariner 9's storage capacity wasn't too much better, but since it wasn't a flyby, it didn't need to take and store so many images in a hurry.
As for Pioneer, what I fear is that the data suffered the fate of the original surveyor data. The surveyor data was preserved in the form of photographic prints, not original transmissions - at the time, afterall, who stored images in an electronic format? I also wonder about the fate of the Pioneer Venus Orbiter Cloud Photopolarimeter dataset. It must be recorded somewhere. But how in the world to convert it into a useable format...
I had not realized the entire set of Mariner 4 matrices were available. That would be a neat project. But a boring one at first.


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

38 Pages V   1 2 3 > » 
Reply to this topicStart new topic

 



RSS Lo-Fi Version Time is now: 25th February 2020 - 09:31 AM
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 a project of the Planetary Society and is funded by donations from visitors and members. Help keep this forum up and running by contributing here.