14
$\begingroup$

JunoCam has tweeted us! More background from the BBC.

enter image description here

It is supposed to have several features that help allow it to work in the high radiation and magnetic fields it will experience during each flyby. Although it has a 4:3 sensor, according to Wikipedia it has a very long, narrow 18 x 3.4 degree FOV, and that there are three color filters, not the normal Bayer arrangement.

How does this thing work? Why such a strange aspect ratio? Inquiring minds want to know!

$\endgroup$
  • 4
    $\begingroup$ The wikipedia article is wrong about the field of view. The camera itself has a 58 degree field of view, but thanks to the pushframe imaging being used, the field of view is effectively 58x360. A much better source: link.springer.com/article/10.1007/s11214-014-0079-x . $\endgroup$ – David Hammen Jul 13 '16 at 10:50
  • $\begingroup$ @DavidHammen Excellent article! I like the part in the abstract where it says "...*the public will weigh in on which images to acquire*..." :-) $\endgroup$ – uhoh Jul 13 '16 at 13:01
18
$\begingroup$

JunoCam used different technologies than does the typical framing camera one buys at a store. A typical digital color camera uses a Bayer filter pattern, a row of alternating tiny blue and green filters, followed by a row of alternating tiny green and red filters, each filter covering a pixel, followed by a row of alternating tiny blue and green filters, and so on. JunoCam instead has four filter strips, each of which is 1600 pixels wide. Three are in the visible range, the standard red, green, and blue filters based on how the human eye works. The fourth is in the near infrared and is designed to see methane. Each of the three visible filter strips are about 150 pixels high; the methane filter strip is a somewhat larger than one of the visible filter strips.

JunoCam uses two other concepts that you don't see in the typical digital camera. One is time delay and integration. The lighting level at Jupiter is about 1/27th of the lighting level at Earth's orbit (Jupiter orbits at about 5.2 AU). Juno rotates at 2 RPM. A short duration exposure would result in too much noise due to the low lighting. A low duration exposure would be too blurry thanks to that rotation. Time delay and integration means taking a series of short duration exposures and integrating them together by shifting the subsequent images to account for rotation.

The other key technique is that JunoCam is a pushframe imager as opposed to a framing imager. A frame results when the time delay and integration is complete (the number of TDI steps is commandable). This frame is pushed onto local memory. JunoCam alternates between visible and methane channel, so that means that another visible frame will be produced a minute later thanks to JunoCam's 2 RPM rotation rate. The result is a jumble of partially overlapping framelets that need extensive ground processing to make sense of the imagery.

JunoCam was designed to yield optimal performance at one hour before and one hour after closest approach. This should give good performance at closest approach, but more importantly should also give good performance over Jupiter's polar regions. JunoCam won't perform so well Juno is far from Jupiter (which is most of its orbit).

$\endgroup$
  • $\begingroup$ Very nice! Thank you for the nice write-up. I had no idea about the NIR filter! $\endgroup$ – uhoh Jul 13 '16 at 13:02
  • 1
    $\begingroup$ Is there any chance you'd consider adding the link to the JunoCam article here also? Then you could cite Figure 12 and also Figure 13 and add include them in the answer. It's an "aha" kind of thing. Definitely not the RGB filters used to simulate human color vision! $\endgroup$ – uhoh Jul 13 '16 at 13:08
  • 1
    $\begingroup$ Nice to see someone well-informed about JunoCam! I've invested some time to help correcting the obsolete Wikipedia paragraph in the German Wikipedia (de.wikipedia.org/wiki/Juno_(Raumsonde)#JunoCam). But the English Wikipedia version about JunoCam (en.wikipedia.org/wiki/JunoCam#Specifications) is still based on an obsolete 2005 technical document. Since Wikipedia is used frequently as a reference, maybe it would be useful to be corrected by a native English/American expert. I've seen the same confusion as in the question several times in forums. $\endgroup$ – Gerald Jul 13 '16 at 20:51
  • 1
    $\begingroup$ ... Btw.: Although the color filters are about 150 pixels high (plus a gap of about 5 pixels), the readout for the framelets is only 128 pixels high, such that for small TDI you don't get a mix of two colors. $\endgroup$ – Gerald Jul 13 '16 at 21:00
  • 1
    $\begingroup$ @uhoh: Thanks for taking this part. I'd think, that Figure 15 is copyrighted by MSSS. The paper as a whole is copyrighted by the authors of the paper. Candice Hansen is JunoCam PI, Glenn Orton (JPL) is usually contact for Earth-bound observations. It's probably best to contact the lead author, i.e. Candice Hansen, with a short request for allowance using the image in Wikipedia: psi.edu/about/staffpage/cjhansen She'll then forward the request to MSSS, if necessary. $\endgroup$ – Gerald Jul 14 '16 at 14:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.