Ok and now for something completly different, i found this today at flickr code blog:
A robot intelligence has invaded Flickr. The “blind astrometry server” is a program which monitors the Astrometry group on Flickr, looking for new photos of the night sky. It then analyzes each photo, and from the unique star positions shown it figures out what part of the sky was photographed and what interesting planets, galaxies or nebulae are contained within. Not only does the photographer get a high-quality description of what’s in their photo, but the main Astrometry.net project gets a new image to add to its storehouse of knowledge. Needless to say this is one of the coolest uses of Flickr groups and the API that we’ve ever seen. I recently discussed the project with team member Christopher Stumm, since he was the one who had the idea to hook it into Flickr.
Your scale and rotation invariant hashing algorithm is fiendishly clever. Where does it come from?
It’s an adaptation of an old idea in computer vision — “geometric hashing” — to astronomical images. It was originally created by researchers who were trying to model associative memory; “that shape reminds me of something I’ve seen before”. The idea works great for astronomical pictures, because stars are easy to locate exactly. By adding a fast search method for similar-shaped arrangements of stars, and a check that eliminates coincidental matches, we’re able to match an image against the whole sky, usually in a matter of seconds.
Some people are so smart, i’m realy jealous 🙂 But still time to take out the camera and start shooting the night sky !