Crack monitoring of masonry walls with standard and enhanced Digital Image Correlation methods

Many contact and optical methods have been developed and tested over the past decades for structural crack monitoring. Contact instruments provide only localized information and require direct contact with the monitored surface. In contrast, optical methods remotely offer information at a larger scale. Among optical methods, Digital Image Correlation (DIC) has been widely used to monitor surfaces by tracking points in images collected using cameras with fixed positions between each collected frame. However, the fixed setup significantly limits the applications for long-term monitoring. To overcome DIC limitations, we investigated Crack Monitoring from Motion (CMfM) which can measure the propagation of cracks over time in images captured using cameras with non-fixed positions. Here, we present the results obtained during laboratory tests on masonry walls by comparing DEMEC mechanical strain gauge measurements, standard DIC and CMfM. Specifically, we measured the crack propagation in masonry walls under compression and bending loading at different stages using the DEMEC instrument and by collecting images with fixed and non-fixed cameras. We processed images collected with the fixed camera using Py2DIC standard DIC software and images from non-fixed cameras using CMfM. Comparing the crack propagation measured with the three techniques showed a high level of agreement, i.e. a few hundredths of millimetres in terms of median, Root Mean Square Error (RMSE) and Normalized Median Absolute Deviation (NMAD) of the differences, demonstrating the potential for monitoring cracks with permanent and non-permanent camera setup.