Juno development, launch, and cruise, Including Earth flyby imaging Oct 9 2013 Options

[mcaplinger]

I put the partial for efb14 on the website. If you stretch it up, there's something there -- maybe somebody who knows South African geography can say what it is.

[Bjorn Jonsson]

Great images posted here so far, especially when keeping in mind that the camera isn't optimized for Earth/Moon imaging.

I *think* I have now figured out how to properly reproject this stuff using the SPICE kernels and thereby correct for 'everything' - the spacecraft's spin and movement while acquiring the images etc. It's going to take some time to completely finish and debug though.

The biggest uncertainty now is exactly how to correct the geometric distortion (caused by the lens/optics) in the images but I'll probably be able to do this with sufficient accuracy.

[Gerald]

So you found out more than me. I've still to struggle with 18 parameters for the color images to beat down distortions to about 2 pixels.
Even the overlap of the strips varies with x. I'm now using a 2nd order approximation for the overlap.
Fortunately the params seem to stay almost the same within an image series.

[mcaplinger]

The biggest uncertainty now is exactly how to correct the geometric distortion...

The last set of simulations on the website were generated using a first-order radial distortion model of the lens and the actual line coordinates of the filter readout areas (slightly different than what's in the instrument I kernel; I didn't get a chance to update that.) Of course the agreement between simulation and reality isn't perfect.

The Junocam lens has low distortion (less than 3% at the field
corners, primarily barrel distortion.) Low distortion was needed so
that TDI could be employed. Because of this, Junocam can be modeled
to first order using a simple f tan(theta) pinhole camera model. This
can be further improved by adding one radial distortion term K1
following the formulation of [Brown, Duane C. (May 1966). "Decentering
distortion of lenses". Photogrammetric Engineering. 32 (3): 444-462]
of value 3.839251e-8. Measured focal length was 10.997 mm.

[Astro0]

My attempt at manually tweaking the images for a pleasing look.

[djellison]

I put the partial for efb14 on the website. If you stretch it up, there's something there -- maybe somebody who knows South African geography can say what it is.

I'm guessing a little bit here - by comparing shapes of Urban areas with this - http://www.google.com/gadgets/directory?sy...id=703857012153

Also attached - METEOSAT IR image from 2000UT on the night of the flyby - fairly clear, but some cloud N/NE OF J'berg. Hopefully that city will be visible once we get a bit more image.

Attached thumbnail(s)

 

[JRehling]

It would certainly be fun if the CH4-band images showed dark spots where large cattle farms exist.

[mcaplinger]

It would certainly be fun if the CH4-band images showed dark spots where large cattle farms exist.

Only about 20% of atmospheric CH4 is from cattle. I doubt that this imaging is sensitive enough to see any terrestrial variation in CH4, which is only on the order of a few tenths of PPM at the surface based on http://en.wikipedia.org/wiki/Atmospheric_methane -- but I haven't done any analysis to back this up.

[dilo]

Totally fanciful and just for the heck of it. Not scientifically accurate...disclaimer, blah, blah, blah blah, etc, etc....

I like a lot, great!

[mcaplinger]

Aviation Week is reporting "NASA's Jupiter bound Juno probe exited safe mode late Friday.
The exit occurred at 5:12 p.m., EDT, according to a statement from the Southwest Research Institute of San Antonio..."

We still have lots of small gaps in our image data that makes decompression problematic. For the motivated among you, I've put an example on our web site at http://www.msss.com/junocam_efb/images/efb12x.jpg -- this is the RGB image with the best geometry and least compression. I'd be curious how well it can be reassembled. Hopefully we'll have all of the data down by early next week; the spacecraft team is taking a well-deserved rest over the weekend.

[Gerald]

A first draft of the efb12 image fragments looks like this:

[mcaplinger]

A first draft of the efb12 image fragments...

Very nice! This confirms my suspicions that there wasn't any region that managed to get full RGB coverage. In areas where you have two colors, it might work to synthesize the third using the old G=( R+B )/2 trick. It'd also be nice to move the bottom to the top and get rid of all the black in between.

Of course only if you're interested; I'm hoping that we'll have clean data in a few days.

[Gerald]

Reconstructing the third color is an interesting idea. I'll try that later this week-end.

First I tried to get a black/white image out of the fragments by channel split and combining:


Stitching the two halves on a 2d-basis doesn't match quite perfectly, probably because the probe was moving during the take.

[Paolo]

Aviation Week is reporting "NASA's Jupiter bound Juno probe exited safe mode late Friday.

I add a link for that story

[Astro0]

I'll leave the RGB work to someone else. In the meantime, here's an attempt at a stitched/blended version using the images available.



Can't wait to see the ultimate versions to come and just imagining what Jupiter will be like a few years from now.