Automatic integrated system for monitoring superficial cracks

The authors have been studying the use of image processing techniques to monitor cracking in quasi-brittle materials during laboratorial tests. So far, the presence of discontinuities in the image other than cracks, e.g. surface imperfections, has prevented the complete automation of the method, thus requiring the intervention of the user. To overcome this drawback, an integrated approach is herein proposed, using both the displacement field obtained by photogrammetry and image processing.

Direct shear tests (DST) were used to test and validate the new method. Critical areas were automatically identified and the corresponding crack pattern characterised from the start until the complete failure of the specimens. It is stressed out that with this method all stored data is easily processed and analysed. The crack pattern and all related measures (e.g. crack length, area or path) can be provided at any stage.