An Enhanced Deep Learning Approach for Tectonic Fault and Fracture Extraction in Very High Resolution Optical Images

Autor: Bilel Kanoun, Mohamed Cherif, Isabelle Manighetti, Yuliya Tarabalka, Josiane Zerubia
Přispěvatelé: Télédetection et IA embarqués pour le 'New Space' (AYANA), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Géoazur (GEOAZUR 7329), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de la Côte d'Azur, Université Côte d'Azur (UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Université Côte d'Azur (UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud]), Luxcarta technology [Mouans-Sartoux] (LCT), Université Côte d'Azur (UCA), IEEE, ANR-15-IDEX-0001,UCA JEDI,Idex UCA JEDI(2015), ANR-17-CE31-0008,FAULTS_R_GEMS,Les propriétés des failles: une clé fondamentale pour modéliser la rupture sismique et ses effets(2017), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Université Côte d'Azur (UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Université Côte d'Azur (UCA)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [France-Sud])
Rok vydání: 2022
Předmět:
Zdroj: ICASSP 2022-IEEE International Conference on Acoustics, Speech, & Signal Processing
ICASSP 2022-IEEE International Conference on Acoustics, Speech, & Signal Processing, IEEE, May 2022, Singapore/Hybrid, Singapore
ICASSP 2022-IEEE International Conference on Acoustics, Speech, & Signal Processing, IEEE, May 2022, Singapore/Hybrid, Singapore. ⟨10.1109/ICASSP43922.2022.9747007⟩
HAL
DOI: 10.1109/icassp43922.2022.9747007
Popis: International audience; Identifying and mapping fractures and faults are important in geosciences, especially in earthquake hazard and geological reservoir studies. This mapping can be done manually in optical images of the earth surface, yet it is time consuming and it requires an expertise that may not be available. Building upon a recent prior study, we develop a deep learning approach, based on a variant of a U-Net neural network, and apply it to automate fracture and fault mapping in optical images and topographic data. We show that training the model with a realistic knowledge of fracture and fault uneven distributions and trends, and using a loss function that operates at both pixel and larger scales through the combined use of weighted Binary Cross Entropy and Intersection over Union, greatly improves the predictions, both qualitatively and quantitatively. As we apply the model to a site differing from those used for training, we demonstrate its enhanced generalization capacity.
Databáze: OpenAIRE