Dione in living fakey color
Dec. 15th, 2004 08:22 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinI tried my hand at assembling a color picture out of separations of Saturn's moon Dione that came back today from Cassini:
The picture nicely brings out the streaked, stained-looking region on the trailing hemisphere. The crater with the bright rays near the bottom appears in Voyager maps and is called Cassandra; the distinct white streak near it with the hairline down the middle is Padua Linea. Even mangled by me, this picture has about as much detail as the best Voyager pictures, which were of the other side of the moon; the raw black-and-white Cassini exposures were a lot sharper.
The color differences are probably mildly exaggerated from reality, since I made no attempt at calibrating the image, and this is assembled from ultraviolet, green, and infrared exposures. Also, I couldn't get the registration exactly right with the tools available to me, since the spacecraft (and the moon) moved in between color channel exposures, so they were all taken from noticeably different perspectives; that's why most of the craters have fuzzy color fringes, and for wondrous clarity you're better off looking at the raw archive for now, which also has closer images.
Update: Improved the picture considerably from the original version, by playing some more tricks, in particular using a B/W picture as the brightness reference and the three-component composite only for hue/saturation information. I'm starting to appreciate what the Voyager teams in the 1980s must have gone through using digital compositing tools that were probably less powerful than GIMP.
I also clipped out the part of the picture on the right that had data dropouts on every other scan line. I don't know what's up with those; it looks as if it's somehow compression- or bandwidth-related, since the dropouts seem to preferentially occur on scan lines that have complex image data on them.
Here's a simulated view from Celestia based on Voyager maps of that hemisphere (as adapted via USGS cartographers into Jens's DDS texture for Celestia):
The crudity of this is mostly due to the fact that the closest Voyager encounter was on the other side. Notice also that the USGS map shows brightness features that aren't visible in the Cassini picture, probably because of differences in illumination angle.
Another update: ...Or maybe not. Comparing the Schenk photomosaic from the Planetary Society page to the USGS map makes me think that the map has hemispheric brightness differences played down in favor of depicting topography, and that those easternmost streaks are a bit fanciful; mapping the actual Voyager photomosaic to the sphere produces something much more like the Cassini photos, albeit at much lower resolution. In any event, it's always nice to be watching when people push a little way into Here Bee Dragones territory.
The picture nicely brings out the streaked, stained-looking region on the trailing hemisphere. The crater with the bright rays near the bottom appears in Voyager maps and is called Cassandra; the distinct white streak near it with the hairline down the middle is Padua Linea. Even mangled by me, this picture has about as much detail as the best Voyager pictures, which were of the other side of the moon; the raw black-and-white Cassini exposures were a lot sharper.
The color differences are probably mildly exaggerated from reality, since I made no attempt at calibrating the image, and this is assembled from ultraviolet, green, and infrared exposures. Also, I couldn't get the registration exactly right with the tools available to me, since the spacecraft (and the moon) moved in between color channel exposures, so they were all taken from noticeably different perspectives; that's why most of the craters have fuzzy color fringes, and for wondrous clarity you're better off looking at the raw archive for now, which also has closer images.
Update: Improved the picture considerably from the original version, by playing some more tricks, in particular using a B/W picture as the brightness reference and the three-component composite only for hue/saturation information. I'm starting to appreciate what the Voyager teams in the 1980s must have gone through using digital compositing tools that were probably less powerful than GIMP.
I also clipped out the part of the picture on the right that had data dropouts on every other scan line. I don't know what's up with those; it looks as if it's somehow compression- or bandwidth-related, since the dropouts seem to preferentially occur on scan lines that have complex image data on them.
Here's a simulated view from Celestia based on Voyager maps of that hemisphere (as adapted via USGS cartographers into Jens's DDS texture for Celestia):
The crudity of this is mostly due to the fact that the closest Voyager encounter was on the other side. Notice also that the USGS map shows brightness features that aren't visible in the Cassini picture, probably because of differences in illumination angle.
Another update: ...Or maybe not. Comparing the Schenk photomosaic from the Planetary Society page to the USGS map makes me think that the map has hemispheric brightness differences played down in favor of depicting topography, and that those easternmost streaks are a bit fanciful; mapping the actual Voyager photomosaic to the sphere produces something much more like the Cassini photos, albeit at much lower resolution. In any event, it's always nice to be watching when people push a little way into Here Bee Dragones territory.