Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025Geophysics

Unsupervised seismic random noise attention using 3D enhanced multi-scale features

View Full Paper
Ask AI
Bookmark
Share

Authors

MZMi ZhangCJChangqing JingYCYongfu Cui

Discussion

Loading...

Member takes

Overview

Unsupervised deep learning improves denoising in 3D seismic data, suggesting novel approaches to noise suppression.

Key Points

  • The EMST method effectively reduces random noise and improves seismic signal quality.
  • Using multiple datasets, the method showed superior denoising performance and minimal signal leakage.
  • An attention mechanism significantly boosts the feature representation at various scales.
  • Monte Carlo patch selection enhances the training efficiency by optimizing representative data patches.

Cite This Study

Zhang et al. (2025) studied this question.

synapsesocial.com/papers/68c19fa854b1d3bfb60db826https://doi.org/10.1190/geo2024-0891.1
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Seismic Random Noise Attenuation via Channel Attention‐Weighted Unsupervised Deep Learning2026
  2. 2Ground-truth-free deep learning for 3D seismic denoising and reconstruction with channel attention mechanism2024 · 11 citations
  3. 3A Hybrid CNN-Transformer Network for Linear Noise Suppression in 3D Seismic Data2026
  4. 4Multi-scale dual-path attention network for seismic background noise attenuation2025
  5. 5Multi-scale dual-path attention network for seismic background noise attenuation2025