I wonder how this compares to purely vision-based systems which use nothing but the images themselves for stabilization. Here are some quite old results of stabilization using image-based 3d-reconstruction of the scene which I wrote more than 10years ago, compared with other stabilization programs of that time (Deshaker, Adobe After Effects, Youtube). With todays improved hardware and progress in 3d-algorithms you may not need any additional gyroscopic data.
"Viewsheds" of any location can be calculated and matched with photographs using "GeoImageViewer", an application I wrote a couple of years ago. Any feature in the photo can be interactively identified in a mapview and vice versa, including the boundary of the viewshed. See the link below for a couple of examples.
I wonder how atmospheric refraction is handled in the calculations for the longest line of sight. Since it (a) strongly affects the line of sight, and (b) depends on temperature and weather, how is a static "world record" possible, or even defined? E.g. objects may appear 400m higher in 200km distance under typical conditions.
Atmospheric refraction is due to the vertical gradients of atmospheric pressure, temperature, and composition of the atmosphere, all of which are usually not precisely known, and which vary with time, so one gets larger lines of sights at certain times. For my application I used the standard formulas for astronomical refraction (-> many weblinks) with constant medium gradients. If I recall correctly this results in ~100m height correction for features in 100km distance and ~400m in 200km distance (features appear higher than without atmospheric refraction). For your application it would make sense to use two extreme values for the gradients to get maximum and minimum, as you suggested.
"Viewsheds" of any location can be calculated and matched with photographs using "GeoImageViewer", an application I wrote a couple of years ago. Any feature in the image can be interactively identified in a mapview and vice versa, including the boundary of the viewshed. As has been mentioned in the comments, it is essential to include atmospheric refraction in the calculation, at least for distances above ~100km.
https://www.youtube.com/watch?v=-m3fwhx3Z5g