Hello! You've an interesting theory, but in practical you've not mentioned how exactly you wish to detect damaged structures/buildings using just 2D information?
I speak with experience of working with LIDAR data for the exact same purpose during cloud bursts few years ago in India.
What you're correct with?
Segmentation of the building and vegetation area, however if considering the geometry and their edge maps, most vegetation area should be eliminated anyway. Okay, with color information, more can be done, things can be slightly better.
Where you're still unclear?
How do you prove a broken edge is in fact a damaged structure? Kindly elaborate more of this part, so that I can get a good idea of the time it will take for the implementation.
About me:
Am a research assistant in image processing and machine learning field, so if you like me to work on any algorithm changes, I can help you with it. However, it will be additional to the implementation costs. Perhaps, we can discuss in more details, if you're interested.
Approach:
On a brief note, we shall divide it into 2 to 3 milestone payments, for update tracking and delivery. Tools: Matlab, Python
Kindly leave your queries and clarifications even if I'm not online, and I shall revert back ASAP. Thanks!