Building surface damage recognition based on synthetic data

To detect surface damage to buildings, it is necessary to involve workers who are at risk of industrial injuries when inspecting hard-to-reach areas of industrial premises. Attraction of special means, such as aerial platforms, safety systems, etc. increase the financial costs with this approach. The use of unmanned aerial vehicles, coupled with neural network algorithms, can simplify this procedure. Due to the inaccessibility, the problem of obtaining training data for neural networks arises, which can be solved by synthesizing them in a virtual environment.