Automatic Crack Detection and Analysis for Biological Cellular Materials in X-Ray In Situ Tomography Measurements

Abstract

We introduce a novel methodology, based on in situ X-ray tomography measurements, to quantify and analyze 3D crack morphologies in biological cellular materials during damage process. Damage characterization in cellular materials is challenging due to the difficulty of identifying and registering cracks from the complicated 3D network structure. In this paper, we develop a pipeline of computer vision algorithms to extract crack patterns from a large volumetric dataset of in situ X-ray tomography measurement obtained during a compression test. Based on a hybrid approach using both model-based feature filtering and data-driven machine learning, the proposed method shows high efficiency and accuracy in identifying the crack pattern from the complex cellular structures and tomography reconstruction artifacts. The identified cracks are registered as 3D tilted planes, where 3D morphology descriptors including crack location, crack opening width, and crack plane orientation are registered to provide quantitative data for future mechanical analysis. This method is applied to two different biological materials with different levels of porosity, i.e., sea urchin (Heterocentrotus mamillatus) spines and emu (Dromaius novaehollandiae) eggshells. The results are verified by experienced human image readers. The methodology presented in this paper can be utilized for crack analysis in many other cellular solids, including both synthetic and natural materials.

Computer‐Vision‐Based Approach to Classify and Quantify Flaws in Li‐Ion Electrodes

“…While this proof of concept segmentation is relatively simple, it opens the opportunities for further analysis of cracks and flaws, for example by studying their internal surface areas and orientations. [ 32 ] Furthermore, the crack severity could be estimated by measuring the distance of the crack to the particle surface. After labeling and classification, the electrode could be spatially reconstructed, and the flaw distribution could be spatially localized throughout the depth of the electrode.…”

High strength and damage-tolerance in echinoderm stereom as a natural bicontinuous ceramic cellular solid

“…The center of each segmented crack plane was identified by averaging all voxel locations in the crack, and the crack plane orientation was determined by the minimum principal component with principal component analysis. Further details regarding crack detection and analysis can be found in a previous work 63 . For the analysis of damage bands at the later stage of deformations, we adopted a U-net architecture in a DNN to learn the features of the fragments from the manual labeling.…”

Using Deep Learning to Predict Fracture Patterns in Crystalline Solids

“…With binary_crossentropy, we can interpret the probability of crack in each pixel from 0 to 1 as the probability of crack on different positions. [48][49][50] Through this approach, the crack prediction problem become a multiclassification problem.…”