NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Position from a photo
From: Bill B
Date: 2012 Dec 27, 15:25 -0500
From: Bill B
Date: 2012 Dec 27, 15:25 -0500
> Suppose you wish to covertly record the location of a stash of gold > buried in the Sahara desert (say). Could you take a casual looking photo > which would convey enough information to make a fix within a few nm of > where the photo was taken? If camera has a gps in it, megadata would have the fix. Wide angle and *a* photo (one photo, as per the stated problem)? As list "camera as a sextant" master Greg has pointed out in the past the pixel counting works best (or easiest at least) with a telephoto using the center of the lens. A wide angle would be, I suspect, a bear to calibrate. Ken G in the past claimed he had a drawer full of images of the sun where he could tell the elevation of the sun by the amount of squish caused my refraction as it neared the horizon. At last mention he has yet to write his book which would include that method. Being desert, abnormal refraction might be major hurdle as well. At this point I'm still scratching my head. Any method I can imagine with one image has major drawbacks, including the need to record time and date either in the megadata or by hand. Is an image with a watch and or calender included a "casual" photo? And just how does one get the depth of field to have both a body at infinity and the watch in focus short of the Lytro light-field camera? I haven't thought it through, but perhaps latitude through elevation of a body via pixel counting or squish and a time sight for longitude? Bill B.