Hi,

I'm working on the Pioneer Jupiter images and I'd like to share this with you. It's a rough cylindrical mosaic of Jupiter.

Even working from imagens scanned from paper modern processing can make wonders to old data!

Click to view attachment

WOW!

I like that!

Great work! I had a few Pioneer images in digital matrixes. I was about to release them before my crash. I had entered them manually. Someone (Bjorn I think) said they had a program that could scan the matrices and convert them to text. My original copies were too bad to scan, but I now have better copies. I would be willing to send the scans someone with the capability to convert them to images on the condition that they share the new "raw" images with me and allow me to process them too (they are also free to do what they want with them - it just took me literally hundreds of hours to find these, so I don't want to lose access). There is a red and blue pair of Saturn's rings, and there is a red channel image of the rings against the disk of Saturn.

Ted

.

The avaliable Pioneer scanned images are not that bad. I'll put them on my site when I have the time. I've been busy color correcting and resampling them to their aproximate original pixel sizes. I'm actually surprised that the images have so little noise, given that the original data is only 6 bits. For some I have individual Red and Blue channel images, that I used to make R(g)B images.

I don't thing that using real digital data would improve things much. Perhaps a good high quality scan of the printed images would do. There are some on the net and you can clearly see the original image pixels, meaning that all the information is preserved.

When I have the time I'll post more images.

I will tell you from my experience that it doesn't. For example, in my Io work, the digital image doesn't show much more than the scanned version - and a lot of the improvement could have been recreated had I simply restacked the scanned red and blue images rather than just scanning a color one.


The print versions are very good. I think a big reason that they appear so noise free is that the 6 bit format was used due to Pioneer's very limited data rate and the fact that all data had to be sent in real (or almost real - the Pioneers had a small amount of storage for data taken during planetary occultations) time. The IPP put out a signal that would have benefitted from a 7 or 8 bit format. Therefore, most noise (along with a lot of detail) was lost in the encoding.

Ted, I'll volunteer to do the typing, if you can't find a scanning program. Since I know almost nothing about processing, I'd hand it right back to you once the file's digitized.

If you're interested, contact my email. (I just turned put it in my profile.) Maybe you could email a scan, or send Xeroxes. (For God's sake don't send any originals!)

Of course, since your hard drive committed suicide right after you loaded it, maybe the data's cursed.

(I was just about to order red & blue Pioneer negatives from the NSSDC, and here comes some digital data. NSSDC reprints go for $10 a pop. Each. Better have a good idea what you want to order!)

John D.

I am at work, so I don't have them now. I can type the numbers in as text. The issue is that the only way I have to convert that to an image is to create an indexed photoshop image and individually select each pixel, entering a value, which takes forever. I will get to work on this when I get home tonight.

I've got a bunch of Perl scripts that will do this (and the reverse - pushing out pixel values into arrays\matrices). There is certainly no reason to do something as painful as this by hand if you have the data it will take me minutes at most provided the data is more or less coherent.

Yeah, I was about to say, "WHAT?" Doesn't your program know about "raw" formats?

I had assumed I was going to type a text file, then write a quick program that would turn that into a binary stream of bytes. (Then take a quick look in Paint Shop Pro loading it as a raw.)

Heck, I bet you could load up a spreadsheet like Excel, fill in the cells with the numbers, and turn it into special 3D graph to view it directly. I KNOW you can do this in Mathcad from version 5 up: create a matrix, then display that matrix as a B/W contour plot.

John D.

Converting a simple text file containing numbers (e.g. one row for each image row) to something like PNG is trivial for me. In fact I already have a program I wrote for this purpose late last century when I was processing intensity profiles of Saturn's rings in text format.

Today I suddenly remembered that I have the Pioneer 11 Saturn special issue of JGR and it contains something interesting: Pages 5918-5919 contain a printed table of numbers for red and blue channel images of the rings. There is another smaller one (red channel only) on page 5920 showing the rings silhouetted against Saturn (Saturn's limb is also visible). I remember typing at least one of these into a computer ages ago but I cannot find it - I probably lost it ages ago when moving stuff between computers.

I wonder if these are the images Ted has in mind. This appeared in JGR vol. 85, no. A11 (November 1, 1980).

BTW there is a related thread here. Maybe these two threads should be merged...

Despite the comments above on the good quality of scanned images there's another reason for wanting this data in digital format: Getting *all* of the Pioneer images (the majority hasn't appeared in print). I have all of the Voyager, Galileo and Cassini images and my collection of outer solar system spacecraft images isn't complete without the Pioneer images .

Yes, I do have those. Here are scans of them.

wow - reminds me of this
http://grin.hq.nasa.gov/IMAGES/MEDIUM/GPN-2003-00060.jpg

Doug

Appendix 2 of the NASA publication Pioneer Odyssey (1977 edition)
contains the technical details of the Jupiter images from the two probes,
if this helps any:

http://history.nasa.gov/SP-349/app2.htm

Here is a view I have worked on from scanned data. It is a "look-back" image. Due to data rate limitations, Pioneer only transmitted every other line of this image. The original is extremely jagged because of that.

Great!

Bob Shaw

Here's an Excel file with part of the "f6fragment9xg" typed in... I'll keep at it, but if someone want's to do it, at least here's part of it ;-) I'm typing colums 0 to 10, row by row (almost done) and then I'll move on to colums 10-20. If someone want's to start at the oposite end (colums 30 to 40) it would be a great help!!!
Click to view attachment

This is possibly the smallest image I'll ever upload here but here's a png version of that data. I've stretched the intensity x 5 to bring out the detail a bit.

Um, I think you have your numbers off. it looks like you're only doing the top left part of the image. Here's my contribution, the right 5 rows (AJ-AN), from (1-60). I also changed the legend so that it is a graduated grey scale instead of the pastel colors.

Click to view attachment

Click to view attachment

Yes, I'm doing the left part of the image ;-)

Here's our combined progress so far... almost done!

Only the middle to go!!!

Click to view attachment

Oddly I have seen very little Saturn data from Pioneer.

Is there a archive images online?

I hadn't seen that one.

Pioneer 11's Saturn encounter was one I recall very fondly. The Newsweek that had one of the images on the cover, I saved for years.

It's about the only flyby mission I can recall (besides the Venus ones without a camera, smart guys) that DIDN'T improve on telescope images of the world, but it was from, by the standards of what had been done before, from so far away, that it was impressive on those grounds alone.

Actually, it did somewhat, such as with the ring images....it certainly broke them down more in the closest views, which are rarely seen. But the slight improvement was not enough to be revolutionary in terms of resolution.

Well, there was one photo that was definitely better -- the famous one which simultaneously revealed the existence of the F Ring AND confirmed the existence of a 10th moon of Saturn (which turned out to be Epimetheus). Also, note that Pioneer 11 did give us the very first eerie photos of Saturn's rings seen from behind.

This is an sythetic view the Pioneer team produced from two images. It clearly shows more ring structure than earthbased photos of that time.

Just to see what it looks like.
Direct conversion to png (pixel RGB intensity values are set to the values in the spreadsheet)
Click to view attachment
Stretched x3 to bring out the contrast a bit.
Click to view attachment

Here is the completed matrix!

Here ya go. Small but sweet
Raw:
Click to view attachment
Stretched x 3
Click to view attachment

Darn it. You guys are way too fast. (I know, that's a good thing.)

Yesterday, I printed out the numbers so I could type them in on the train on the way to work. I had three columns done, double-checked. I get to work, and discover it's half-done!

Of course, the next morning (now), it's all done.

Oh well. Here's some more formats:

1. f3.csv: Comma-delimited text file (easy for loading in other programs.)
2. f3.raw: RAW binary file, exactly 60 * 40 bytes long.
3. f3scaled.raw: RAW binary, with values scaled from 0-63 to 0-255. (That's 4x, not 3x)

I had wondered at the "-1"s in the values, and noticed that someone else had decided that was dropped data and averaged. Also one column was changed from a duplicate to an average. In the interests of historical accuracy, I made a zeroed version so everyone can see the original blank lines:

3. f3raw.raw: RAW of the original "raw" pixels. (OK, that's a confusing name.)
4. f3rawscaled.raw: values scaled to 0-255

(Funny, if you drop the empty column, the rings arc looks more lined-up.)

...Never mind. Now I find I can't upload files with extensions ".csv" or ".raw" Ted, I'll email them to you directly. Everyone else, I'd give them out if I could find a place to put them!

I'll start on the bigger red & blue matrices, but I have a feeling I won't be able to catch up to all of you.

I think our speed had to do with several of us working on it. Great to finally see it as an image.

Click to view attachment

Here's a magnified Gif version of the image, oriented as the printed data (planet limb/terminator? at top left) stretched and gamma corrected.

Yes, I replaced the -1 values with averaged data, and did the same for a duplicated column :-)

Click to view attachment

And here's the red rings image typed in

Rest assured, I'll not type the blue one so soon so if someone wants to do it go ahead!!!

Here's the PNG for your troubles at least. Stretched x4 as that definitely seems to be correct and oriented correctly this time.
Click to view attachment

I am at work on the blue image.

Ted's probably already finished with this, but for completeness, here's my Excel version of the Blue-channel rings.

Note that there may be an offset from Helvick's version. Mine is 73 x 86 ("BU" across), and I included all the "0"s on all sides. (I think Helvick skipped the leftmost column of zeros.)

Click to view attachment

I see your point. The data for the red image seems to be offset horizontally by 1 pixel (to the right) when compared to your blue data. I've tried to bring them into line by adding an extra column of empty pixels to the blue data to bring them both to 73 pixels across.

Anyway here's some png's:
Red:
Click to view attachment
Blue:
Click to view attachment
and a color version using both of these and a synthetic green channel.
Click to view attachment

I am glad you did it...I still am not finished

Ted

Here's a quick processed version, with colour data merged to reduce noise. Colors were adjusted to a neutral yellow gray and saturation increased 50% to reveal hue variations.

Not bad at all Wish we had more old data to play, even if all of it had to be hand typed! Click to view attachment

I have done some work on it, and this is what I have come up with so far. It is an R-Synthetic Green-B color image merged with a super-resolution attempt with the two images. The F-Ring is artificially brightened.

I have several leads...If I come up with anything, I will post it.

.

Here are the best Pioneer views of the Galileans I could make. Europa and Callisto are entirely from scans. Ganymede is from a mix of scans and some limited digital information taken from a multitude of papers. Io is purely digital.

Is Pele's plume there real or did you add that to make me jump?

Yes, is this the discovery of a pre-discovery volcano image?

Would a plume even be detectable with exposures optimized for Io's dayside and with probably not-so-sensitive equipment? We're talking about the fuzz at the limb, right?

True, though at least it is a somewhat reasonable phase angle to start to get a brighter plume. Do I see a hint of a bluish color as well?

A quick'n'dodgy enhance of the image, I hope Ted doesn't mind:

It indeed does appear to have a plume-like characteristic. If it really is Pele's plume, I'm officially in awe of the sensitivity pulled out of the data. Either that or the eruption was absolutely massive.

..stunning! Same for Ted's montage, esp. the Ganymede image. Amazing that such detail can be extracted from this heritage data; you guys are something else.

We will never know. It is so dependent on so few pixels that while it is tempting, there is no way to be sure whether or not it is a plume or noise, unless there is some other corroborating data, such as infrared photometry, which might indicate the plume was active at the time, which might solidify it. When I first processed the Io image, I sent it to John Spencer, who had the same conclusion - it is intriguing, but can't be proven in and of itself.

Extremely interesting even though I doubt the Pioneer IPP instrument was sensitive enough to detect a plume - maybe it could detect a huge Tvashtar-like plume though.

As a check I enhanced all of the Galileans in Ted's image to check if the other Galileans exhibited similar behavior. They didn't.

I was unable to determine the exact viewing geometry using SPICE kernels but know that the subspacecraft latitude is +60° and the subsolar longitude 184°. So this is the approximate viewing geometry:

Click to view attachment

And without the grid:

Click to view attachment

From these renderings I doubt this could be Pele's plume - it would be closer to the terminator. But needless to say this might be a different plume...

BTW if anyone knows the exact subspacecraft longitude I'd like to know - the source of my geometric information (Pioneer: First to Jupiter, Saturn and Beyond - NASA SP-446) only mentions the subspacecraft latitude.

EDIT: The pixel size in the Io image is ~375 km so apparently the plume's 'source data' only consists of about three pixels.