AUTHOR=Shi Yongxiang , Ballesio Marco , Johansen Kasper , Trentman Daniel , Huang Yunsong , McCabe Matthew F. , Bruhn Ronald , Schuster Gerard 

TITLE=Semi-universal geo-crack detection by machine learning

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2023.1073211

DOI=10.3389/feart.2023.1073211

ISSN=2296-6463

ABSTRACT=Cracks are a key feature that determines the structural integrity of rocks, and their angular distribution can be used to determine the local or regional stress patterns.   The temporal growth of cracks can be monitored in order to predict impending failures of materials or structures such as a weakened dam.  Thus, cracks and their spatial-temporal distributions should be automatically  monitored for assessing their structural integrity, the associated stress patterns and their potential for failure.  We now show that the U-Net convolutional neural network (CNN), semantic segmentation and transfer learning can be used to accurately detect cracks in drone photos of sedimentary massifs. 
  In this case, the crack distributions are used to assess the safest areas for tunnel excavation.  Compared with ridge detection by the shearlet transform, the proposed approach yields fewer false positives and higher accuracy.  The U-Net accuracy in identifying cracks in images is 98% in comparison to human identification, which is sufficient for assessing the general crack properties of the rock faces for the engineering project.  After 20 hours training on 127 photos from 100 hours of labeling work, our network was able to successfully detect cracks in 23,845 high-resolution photographs in less than 22 hours using two NVidia V100 GPUs.  Meanwhile, The network was able to detect more than 80% of the observable cracks of a volcanic outcrop more than 8000 miles away in Wyoming without additional training.   With a modest amount of extra labeling on photos of the volcanic outcrop and transfer training, we found that the accuracy significantly improved.  The surprising outcome of this research is that the U-Net crack detector laboriously trained on photos of sedimentary rocks can also be effectively applied to photos of volcanic rock faces.  This can be important for real-time assessment of geological hazards and lithology information for dam inspection and planetary exploration by autonomous vehicles.  For another application, we accurately detected fractures and faults with a scale of tens of kms on martian photographs, our methodology of using CNN with transfer training suggests that it can be used as a semi-universal detector of cracks in geological bodies.