----- Original Message ----- From: "Björn Gimle @GlocalNet" <bg_26934@glocalnet.net> >For a simple lens, all three factors may be near cos(angle), so raise this to a >power of three! >So these contributions should not be subtracted, but divided! >For a 135 film equivalent focal length of 35 mm I find 31.7 degrees to the >corner, and cos^3 = 0.615 I didn't think of that but now that you mention it you can fix this more accurately of course with a "flat frame" - a picture of gray paper out of focus. But if you don't have access to the camera , you should instead be able to use the existing picture and blur it severly (say 15 pixel gaussian blur maybe). Then grab the light strip from the same part of the image as before and use excel or something to divide the two lists of numbers. Shouldn't be hard and you don't have to get it perfect to improve the results. >Also, satellite speed and distance can vary substantionally across the image. Wow. I thought of distance but not speed. Speed will affect the brightness recorded by the pixels big time. I suppose that both could be factored out if you know the observers location and the orbital elements and you generate a table of angular velocity and distance for the pass. I was also thinking you could look up some of the stars to find out their magnitude and then you would be able to use this to make a magnitude scale. This all sounds like an awful lot of work and the graph is actually quite useful as it is. - George Roberts http://gr5.org ------------------------------------------------------------------------- Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive: http://www.satobs.org/seesat/seesatindex.html
This archive was generated by hypermail 2b29 : Sun Jan 13 2008 - 14:47:49 UTC