rotating satellite imagery can help terrain recognition

Hi all!

I’ve done a fair amount of looking at satellite imagery and tracing stuff on the images. And with most satellite images on screen I have a problem: I just don’t get the visual clues to get a 3D impression of the terrain. I think I’ve found out why that is by playing with satellite image printouts. If I hold the print roughly upside-down, usually I instantly get a fairly good 3D impression. Now why would that be? I don’t know either, but there must be a reason for the hillshades in topographical maps being done as if the light was coming from the north-west, even though the sun usually doesn’t shine from there except for a few very unusual situations. It seems to be the lighting which is easiest for the eye to get clues from. Try it out for yourselves - you don’t actually have to print anything if you have google earth; they have the rotate feature and you’ll see what I mean.

Now some editing for OSM relies heavily on satellite imagery, and being forced to do it north-up and with the sun usually somewhere between SE and SW is in my opinion a pain in the neck (from twisting it so as to get the 3D clues :wink:

I wonder if it would be technically feasible to add a rotate map feature to JOSM etc.? I think editing would be much easier that way.

I guess the technology will improve further in the future and with more better satellites the 3D impression will get better, even without further manipulation.

3D Terrain is great in modern Avionics for the aviation industry too, like the Synthetic vision by garmin, I can see this technology being relevant for other applications too. Seems very sophisticated despite being fairly new.