User:Sennheiser/irc
|
|
<slinted> if you are really interested in how they do color balancing, i could send you a couple links that explain the likely exposure times of the images at differant filter positions, and some human perception graphs as well <sennheiser> yes i am interested <sennheiser> especially since they use infared <sennheiser> are you still there? <slinted> sorry <slinted> one sec, and i'll send you the links <sennheiser> thanks <slinted> first thing to consider is that as they take pictures using filters further down the "blue" end of the spectrum, the exposure times go up, so the images appear at the same brightness, even if they shouldn't be (if an image is all red, and little blue, they leave the camera exposing longer on the blue, to get more detail, but that makes it look relatively brighter) <slinted> http://europa.la.asu.edu:8585/PGG/greeley/courses/pdf/bell_2003.pdf page 93 shows that, as long as 10 seconds for an L7 pic, whereas L2 might be less than a second <sennheiser> thank you <slinted> we don't actually know the exposure times yet <sennheiser> keep sending the links, i have to go away for a while. ill be back soon <sennheiser> sorry about that <slinted> so we can't reconstruct it perfectly, though it'll be included in the PDS <slinted> also, to make it as a human would see it, you need human perceptivity funtions: http://cvision.ucsd.edu/lumindex.htm <slinted> also: http://robotics.jpl.nasa.gov/people/rwillson/papers/2003JE002077.pdf <slinted> if you didn't see that before <sennheiser> thank you <sennheiser> this looks very interesting