More detailed information on how the Callisto map was created and its accuracy

This section contains more detailed information on how the map was created. It is somewhat technical and may be difficult to understand (perhaps incomprehensible !) unless you have some knowledge of the images returned by the Voyager and Galileo spacecraft. Before you read further, it should be noted that this map is intended for use in 3D/animation work; I do not completely rule out that it might be scientifically useful as well but that was not my goal. You have been warned ;-).

For creating this map I used 18 Voyager 1 images, 8 Voyager 2 images, and 46 Galileo images (all the G2, G7 and G8 images and 4 C3 images). In addition, I used one global color image from each of the Voyager spacecraft to colorize the map. These color images were composed from images taken through the spacecraft's orange, blue and violet filters so I actually used 6 images to colorize. Following this some areas were still not colorized; I fixed this by "cloning" the color of adjacent areas. Areas with cloned color are not indicated in the mask map; they are small and this is not significant since the color is probably not very accurate anyway, in particular I am sceptical of any color differences between the Voyager 1 and Voyager 2 images. In summary, the above means that a total of 78 (!) images were used. All of these images were "raw" images from the spacecraft. In addition, I took a small area from a USGS 4 km/pixel map of Callisto.

Positional and photometric accuracy
My original version of the map has a resolution of 4 km/pixel (3784x1892 pixels; I do not have sufficient webspace to put it here). It was created from scratch by reprojecting the individual spacecraft images to simple cylindrical projection, compensating for the varying illumination on the fly. Following this I made 6 (cylindrical) mosaics from the individual "datasets" (Voyager 1, Voyager 2, G2, C3, G7 and G8). It was necessary for me to calibrate the Voyager images using flat field and dark frame images taken for calibration purposes. The reason is that the raw, uncalibrated Voyager images have poor photometric accuracy, for example they are very obviously brighter near the corners than near center. Calibrating the Galileo images was not necessary for my purposes. I also made no attempt to correct for possible geometric distortion of the raw images.

For the Galileo images, reprojecting to simple cylindrical projection was straightforward since the raw image files contain information on the viewing and lighting geometry. The fact that this information is not always 100% accurate created only minor problems. For the Voyager images things were more complicated since no information on the viewing or lighting geometry information is available (or so I think; if this statement is wrong I would like to know !). In this case I used a 4 km/pixel "Voyager-only" map from the USGS to determine the latitude and longitude of several points in each Voyager photo. This makes it possible to determine the viewing geometry and in a perfect world the accuracy of this procedure would be 100%. Unfortunately, my lat/lon measurements are not perfect and there is probably a minor geometric distortion in the images as well, most pronounced near the corners. The USGS map is probably not positionally perfect either. In addition I used David Seal's Solar System Simulator to determine the (approximate) lighting geometry. Despite the above sources of inaccuracies the positional accuracy of my map compared to the USGS map is 5 pixels or better in lat/lon in most cases where this could be determined. For the Galileo G7 and G8 data, comparison to the USGS map was difficult since that map is very fuzzy in the areas covered by G7 and G8. In addition, the longitudinal difference measured in pixels is greater near the poles for obvious reasons.

After creating mosaics from the 6 datasets (V1, V2, G2, C3, G7, G8) I "manually" adjusted the brightness of individual areas within the mosaics using Photoshop since it was impossible for me to compensate to 100% accuracy for the varying illumination of the different images when reprojecting them. I then merged the 6 maps into one big map and removed the seams between them, again using Photoshop (mainly the rubber stamp and gradients for those of you who know Photoshop).

The final step was to add fictional data to areas for which no photos of decent resolution are available. I did this by "cloning" other areas. The appearance of these areas in the map is therefore fictional except for one thing: In some cases it was obvious from the original, very low resolution data where really big craters were present. In these cases I cloned big craters so the location (but not appearance) of big craters in the fictional areas should in several cases be approximately correct.

As the above may suggest, I paid rather little attention to photometric accuracy and doing so probably would have been almost impossible for me anyway. Despite this I am not aware of any really gross photometric errors. The areas near the left and right edges of the map look somewhat brighter than other areas; this may or may not be correct. Also I do not know if there should be a difference in brightness between the polar and equatorial areas; such differences, if any, are not obvious in global images of Callisto.

After this all I had a big grayscale map and all that remained was to colorize it.

Color
For colorizing the map I used two low resolution color images composed from orange, blue and violet images. The color variations within the areas covered by each of them should be real, e.g. the central Valhalla region is certainly not of the same color as nearby areas. In contrast, I am very  sceptical of any color differences between the Voyager 1 and Voyager 2 data. Also some areas use "cloned" color since the two images do not completely cover Callisto's surface. The biggest color gap was near 270 longitude and near the poles, especially the south pole. The reason I didn't use higher resolution color images is that they cover a smaller fraction of the surface than the low-res ones and more importantly, no "reasonable" filter combination is available (the filter selection used for the hi-res images looks rather weird to me but I am no expert here).

I do not know how accurate the resulting color is. In fact I have only a rather vague idea of what Callisto's true color looks like. Most or all of the published Voyager color images are probably false color images or at least have a color that is too saturated. When composing the color images I simply adjusted the R/G/B channels until the color looked reasonable. From the above description the main reason that Valhalla's color in my map differs considerably from the color it has in a well known Voyager 1 hi-res color image of that area can be seen: The filter combination I used for the low-res color images is very different from the combination used for the hi-res image.

Comparison to the USGS map
First the reason why I created this map from scratch instead of simply modifying the Voyager-based USGS 4 km/pixel map: That map has lots of ugly seams. Also it has some gaps/fuzzy areas where better data is available. I suspect this may be in order to ensure positional accuracy, in many cases it seems the parts of the images close to their edges are not used for the USGS map, even though doing so would eliminate/reduce some gaps. This makes sense if there is more geometric/radiometric distortion near the image edges. In some cases (notably east/northeast of Valhalla) smeared images were used - probably smeared because radiation "confused" Voyager's computer. I found different images to use instead of the smeared one(s). There are also several examples where using somewhat lower resolution images than used in the USGS map is actually better, the main case being when specific features are significantly closer to the limb in the higher resolution image than in the lower resolution one.

One of the facts the above description reflects is that my goal was 3D rendering so it was important to minimize the size of gaps and fuzzy areas, even if this meant less positional accuracy. Last but not least my map includes lots of Galileo data which the USGS map lacks (it is based on Voyager imagery only).

As mentioned the difference between my map and the USGS map in the position of features is usually 5 pixels or better. The following table shows several semi-random examples.

Feature Coordinate (USGS map) Coordinate (my map)
Dark spot in Valhalla (3207,771) (3205,770)
Crater in Asgard (2290,632) (2291,633)
North of Adlinda (3558,1289) (3557,1289)
Crater near the north pole (2724,118) (2724,117)
Crater very near the north pole (3550,54) (3548,48)
Crater near the equator (1669,911) (1673,909)
Near an equatorial crater (471,872) (472,872)
Bright creater rim in the south (216,1275) (215,1276)
Crater north of the equator (1793,717) (1788,713) and (1798,716) *

* A nasty example where two images overlap in my map and because of inaccuracies in my determination of the viewing geometry for these two images this particular crater ends up double in the map.

Back to Callisto page