30th Anniversary of the Voyager 1 Flyby of Jupiter Options

[volcanopele]

Today marks the 30th anniversary of Voyager 1's flyby of Jupiter. On March 5, 1979, Voyager 1 flew through the Jupiter system providing a wealth of information of Jupiter, its magnetic field, and moons. Thanks to Voyager 1, the Galilean satellites became worlds with real geology and amazing vistas. Voyager 1 also revealed Jupiter's ring system and Io's volcanism for the first time.

I've written up a longer post about the encounter with Io on my blog, which also has an animation of the flyby: http://gishbar.blogspot.com/2009/03/30th-a...1-flyby-of.html

[ugordan]

I've been experimenting with CIE XYZ colorspace calculations recently with various spacecraft data. This is the same approach I use with Cassini VIMS to generate color, except with camera filters you get many fewer wavelength datapoints so the end result is less accurate. I currently just linearly interpolate the spectral curve between missing points, but it still works pretty well.

As official Voyager images of Jupiter are known to be all kinds of funky colors ranging from greenish, to deeply red I decided to give it a whack. Unfortunately, I don't know how to properly calibrate Voyager images (if anyone has any pointers I'd be glad to hear) so I used raw images off the shelf. The images are c1524909.imq, c1524907.imq and c1524905.imq, OGB frames. Here's a crop of what the straight-up substitution for RGB image channels looks like:



Notice the uncalibrated dark current/bias in the upper left corner, also some color shifting near Europa. The colors are similar to some Photojournal images (albeit they are at a higher contrast stretch) suggesting I at least got the relative channel brightnesses right. Here's the image passed through the code that compensates for the differing wavelengths, and cleaned up in Photoshop:



Compare to a Cassini image processed the same way:

[Bjorn Jonsson]

The cleaned up/color compensated version looks great. The color in many of the ancient, 'official' color images is really bad and they really need to be reprocessed.

I've been experimenting with CIE XYZ colorspace calculations recently with various spacecraft data. This is the same approach I use with Cassini VIMS to generate color, except with camera filters you get many fewer wavelength datapoints so the end result is less accurate. I currently just linearly interpolate the spectral curve between missing points, but it still works pretty well.

As official Voyager images of Jupiter are known to be all kinds of funky colors ranging from greenish, to deeply red I decided to give it a whack. Unfortunately, I don't know how to properly calibrate Voyager images (if anyone has any pointers I'd be glad to hear)

I really found out 'by accident' how to calibrate Voyager images several years ago: I installed ISIS and following that I suddenly had the Voyager calibration files. I then used information from these files (and possibly some of ISIS' source code - I don't remember) to incorparate Voayger calibration into my software. To simplify things I converted most of the calibration files (flatfields etc.) to 800x800 'raw files' containing 4 byte floating point numbers.

By the way - wasn't there a thread somewhere describing in detail the approach you used for generating Cassini (VIMS) color images? I'm just starting a major Voyager image processing project of mine and I'm currently deciding exactly how to process the color images. I might use a similar approach for the first time.