Is attention all geosciences need? Advancing quantitative petrography with attention-based deep learning

• Published in: Computers & geosciences Vol. 181; p. 105466
• Main Authors: Koeshidayatullah, Ardiansyah, Ferreira-Chacua, Ivan, Li, Weichang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2023
• Subjects: algorithms; Attention; Deep learning; geophysics; image analysis; lithology; Microscopy; Petrography; porosity; quantitative analysis; signal-to-noise ratio; Super-resolution; vision; Deep learning; Super-resolution; Petrography; Microscopy; Attention
• ISSN: 0098-3004
• DOI: 10.1016/j.cageo.2023.105466

Recent advances in deep learning have transformed data-driven geoscientific analysis. In particular, the adoption of attention mechanism in deep learning has received considerable interest and shown impressive results that can surpass traditional convolutional neural networks (CNN) in image analysis...