I looked into this sort of system a few years ago using off the shelf equipment ( I am an R.C. flier) I got the cost up to the neighbourhood of $4000. and decided that the return on investment did not justify the cost.
Add to that the possibility of vibration blurring the still pictures ( R.C. engines are notorious vibration generators), exhaust gases blurring the lenses ( all R.C. helicopters blow the exhaust down and that is where the camera is located), vibration destroying the shutter mechanisms of the cameras, the relatively low ‘payload’ capacity of the helicopter requiring the use of small cheap cameras, the need to get really close to the subject ( a zoomed picture is even more effected by vibration) and the VERY STEEP LEARNING CURVE :shock: for flying any R.C. aircraft let alone an R.C. helicopter and you have a very demanding and difficult system to put together and operate.
The use of commercial and military technology could alleviate many of these problems except for the cost which, I believe, will be well in excess of my original estimate of $4000.
None of which answers a very important question; “Should you report on any aspect of a structure that you cannot view directly?” The variables inherent in the image of a remotely viewed roof, using this sort of system are a poor substitute for actually viewing the roof directly.