Applying automated damage classification during digital inspection of structures

With an ever-growing building stock that is showing signs of age-related damage, the inspection of structures is more important than ever. Digitized inspections can help speed up and facilitate the current analog process. The generation of digital twins from building information models with detected and located damage further accelerates the process. Recognizing thereby depicts the classification and localization which is made possible by the deployment of convolutional neural networks (CNNs). This work focuses on the development and deployment of CNNs for live image classification of damage with an Android application in the context of digital inspections.