EPFL researches new crack detection for concrete
Researchers at the Swiss Federal Institute of Technology in Lausanne (EPFL) are developing new methods for detecting cracks in concrete structures. The aim is to use artificial intelligence to create a comparative algorithm that can be used to identify dangerous cracks.
Cracks in reinforced concrete structures are normal and are regularly checked using conventional methods. However, according to a press release, researchers at the Swiss Federal Institute of Technology Lausanne(EPFL) are currently working on methods to digitise these inspections and verify them using an algorithm. Hugo Nick, a Master’s student at the Laboratory for Structural Concrete (IBETON), has developed two methods for this. In the first step, concrete structures are reconstructed in a laboratory and digitally photographed under load until a resulting crack causes the structure to break. The data set is saved and processed into an algorithm using artificial intelligence.
In the second step, digital images of real buildings are taken. These images of cracks occurring in nature and under load will be compared with the algorithm. The researchers hope that this will enable them to draw conclusions about crack behaviour and predict possible hazardous situations.
“A detection algorithm then analyses the image with the help of artificial intelligence,” Hugo Nick is quoted as saying in the press release, “The algorithm is actually a neural network that has been trained on thousands of images and can predict crack detection. Inspectors are beginning to test this method in the field, and it has several advantages. By using automatic crack detection systems, we can reduce the number of errors and inaccuracies and make inspections faster and more accurate.”