Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
After an earthquake, the residual stiffness and strength of structural elements are typically estimated based on a qualitative visual inspection of cracks that is prone to error. In this paper a new approach is proposed to automatically estimate the updated stiffness and strength of damaged unreinforced masonry walls by characterization of crack patterns by a mathematical index. It is shown that structural and textural fractal dimensions of a crack pattern reflect the extent of cracking and the type of cracking or crushing, i.e., whether the cracks pass through joints or whether bricks have been damaged and crushed. Using results of six quasi-static cyclic tests on unreinforced brick masonry walls with different failure modes at various drift ratios, it is shown that the structural and textural fractal dimensions of the crack patterns are an accurate predictor of the stiffness and strength degradation. This procedure seems therefore a promising approach for replacing the traditional assessment methods that are based on visual inspection and engineering judgment by an analytical procedure based on image processing that can eventually be completely automated.
Katrin Beyer, Savvas Saloustros
Aurelio Muttoni, Qianhui Yu, Miguel Fernández Ruiz