Automated 3D Crack Detection for Analyzing Damage Processes in Concrete with Computed Tomography

Analyzing damages at concrete structures due to physical, chemical, and mechanical exposures requires the application of innovative non-destructive testing method combined with 3D image processing algorithms. Since manual segmentation is too time-consuming, we need automatic segmentation methods. In this work, two methods to automatically detect cracks in computed tomograms of concrete specimen are presented and compared. These methods were tested on several datasets of size up to 2000 voxels. Furthermore, the question is addressed of whether automatic crack detection can be used for the quantitative characterization of damage processes, such as crack area and thickness. To achieve such statistical evaluations, a method is proposed to investigate the embedding of the cracks in their surrounding material. Moreover, the representation of cracks as geometrical objects is introduced.