Assessing the structural performance of ageing masonry infrastructure is a complex task. Geometric characteristics and the presence of damage in masonry structures may influence greatly their rate of degradation and in-service mechanical response. Therefore, identifying approaches to assess the actual structural condition of these assets is vital. In the last ten years, advances in laser scanning and photogrammetry have started to drastically change the building industry since such techniques are able to capture rapidly and remotely digital records of objects and features in points cloud and image format. However, the direct and automatic exploitation of images for use as geometry in high fidelity models for structural analysis is limited. In this framework, the aim of this paper is to present the development of a software able to fully automate the “scan to structural modelling” procedure for the efficient and accurate structural assessment of ageing masonry infrastructure. “Image2DEM” is based on Python libraries with graphical interface. The images can be captured from DSLR (Digital Single-Lens Reflex) cameras, smartphones, or drones. The image selected is then imported to the programme to detect and extract the masonry micro-geometry. The algorithm provides reliable detection using Artificial Intelligence. Convolutional Neural Networks (CNN) are used to identify the location of masonry units and cracks, with ∼96% and ∼80% accuracy, respectively. The geometry is extracted in the form of simplified lines to improve efficiency and reduce computational effort. The output is provided in DXF format for compatibility between different programmes. Finally, the geometry extracted is converted to a numerical model for structural analysis. The proposed software has the potential to revolutionize the way we assess existing masonry infrastructure in the future.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)