.png){kind=icon}
.png){kind=icon}
Return
URAL: Uncertainty-driven Region-based Active Learning for data-efficient fault interpretation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cageo.2026.106225.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Pioneering active learning for field seismic fault interpretation. • URAL: uncertainty-driven framework combining sampling and coordinate-aware K-means for optimal regions. • Automatic cropping strategy increases fault-to-background ratio and mitigates sparsity. • URAL matches full supervision with only 3.7% data, showing high efficiency.
Journal
C
IF:
4.4
Papers:
84
Citations:
0