As far as I know, LORRI is panchromatic (black-and-white) and Ralph has red, blue and two infrared filters but no green filter. As such, I wonder whether New Horizons is capable of making natural color images.


Not really, but by being really careful one can get a reasonably close approximation. There are a number of cameras on board, with different wavelengths. The ones that cover visible or near visible wavelenths are (Source):

  • RALPH "Blue" channel 400-550 nm.
  • RALPH "RED" channel- 540-700 nm.
  • RALPH "Pan" (Clear filter)- 450-1000 nm
  • LORRI- 350-850 nm

Okay, so what qualifies as "Natural Color"? I'm going to say that is best done close to the human eye response. Wikipedia has a table with these response values:

  • S 400–500 nm
  • M 450–630 nm
  • L 500–700 nm

Note that there is considerable overlap between the S and M. These are basically blue/green. Given all of this, RALPH can produce a good 2 color approximation in human wavelength, but not really all 3 colors. Some good approximations can be made, however, if one really takes the time to get it close.

Why not natural color? The colors that were selected were chosen to give the most interesting scientific results. Near infrared gives some very interesting information that can't be picked out in visible. Also, the blue overlaps so much with the green that there is little new information by including both bands.

Still, even most color imagers don't quite get the color right. I won't post the image (Not sure about the copyright), but this site contains a the spectral response of the Canon 40d camera, a moderately high end color digital camera. It sees far beyond human vision cutoff (700nm), although in a camera this is typically filtered out at the sensor level with an IR filter. Still, few people would doubt that this camera produces near true color images.

Bottom line, New Horizons doesn't have a "True Color" image, but it can make some pretty close approximations.

See also:

  • $\begingroup$ How it can do close approximations without green channel? $\endgroup$ – Anixx Jul 16 '15 at 17:15
  • 3
    $\begingroup$ There are a few tricks to estimating, using the other bands and subtracting, or simply removing data. But all photos really are some kind of an approximation, so... But honestly, I just know that it would be complicated enough it still wouldn't be exactly what a human mind would see. If you make it out there, perhaps you could let us know;-) I say this having seen very similar camera systems that had simulated natural color images, I personally don't know how it's done, but I know it can be done. $\endgroup$ – PearsonArtPhoto Jul 16 '15 at 17:29
  • 3
    $\begingroup$ There are 4 included sources, plus pulling background from 4 college classes I've taken on the subject (2 graduate level), and a year working with a space mission. I don't know every detail of how these things are done, but I have seen it done on similar missions. The methods are always approximate, but it is entirely possible to do something that approximates reality. $\endgroup$ – PearsonArtPhoto Jul 16 '15 at 17:56
  • 3
    $\begingroup$ @Anixx a natural color range is normally created in color digital camera sensors by assembling information from red, green, and blue sensor pixels. If you have the image with a blue filter, and a red filter, and with no color filtering (which is all the light), then if you subtract the red information and the blue information, then what is left over is the green information. A program can go pixel by pixel and do that. $\endgroup$ – kim holder Jul 16 '15 at 18:01
  • 1
    $\begingroup$ @Anixx that could stand well as a separate question. I admit i noticed that, but i was working on something and didn't have time to look into it, and i let it go. (I have a guess about how much the light on the edges of the filter range gets through compared to the middle, as well, but so what.) Are you interested in asking about it separately? Because i'll post that question if you aren't. Comments don't get read so much and are often deleted once they are old, i really think it deserves a new question. $\endgroup$ – kim holder Jul 17 '15 at 14:00

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.