Finding a camera's Euler angles on earth

[tec]

Hello Everyone,

I would like to find my camera's pointing vector using the stars. Can I determine my camera's pointing vector if I dump an image from my camera onto my computer? Then I could match all the stars brighter than 5 magnitude using a nearest neighbor database. The camera would be about a 8 degree FOV to get greater that 10 stars in the FOV.

I am trying to align my camera and get the pointing vector Euler angles. I could first pick out the star in the middle of the image. Then find that star's RA, DEC angles that the camera is pointing then I could transform them to ECI, then to NED eulers.

I would use Celestia to dump the star data base at a specific time. Then generate a nearest neighbor database. I have used the nearest neighbor data base before to find space satellites's orientation.

The nearest neighbor data base consists of a star of interest with a magnitude brighter than 5, and the six closest stars to that star of interest. Angles would be generated between the star of interest and the neighbor stars using the dot product. This data base would be searched after finding the same angles on the camera image. The star in the nearest neighbor database that matches closest to the camera image is considered the one that the camera is pointing to. I could use weighting factors, too.

Does anyone see a problem with this approach?

Thanks
Tim

[selden]

Tim,

In principle, I think one could write Lua (CelX) functions to do what you want.

Have you investigated Iris? It's intended for analyzing astronomical photographs.
http://astrosurf.com/buil/us/iris/iris.htm