Segmentation of Seismic Images

Abstract

In this paper we propose a method of seismic facies labeling. Given the three-dimensional image cube of seismic sounding data, labeled by a geologist, we first train on the part of the cube, then we propagate labels to the rest of the cube. We use open-source fully annotated 3D geological model of the Netherlands F3 Block. We apply state-of-the-art deep network architecture, adding on top a 3D fully connected conditional random field (CRF) layer. This allows to get smoother labels on data cube cross-sections. Pseudo labeling technique is used to overcome training data scarcity and predict more reliable labels for geological units. Additional data augmentation allows also to enlarge training dataset. The results show superior network performance over existing baseline mode.

Seismic Stratigraphic Interpretation Based on Unsupervised Validation and Spectral Clustering Sampling

“…Seismic exploration specifically utilizes the principles of wave propagation to study subsurface structures and material properties. By studying the intricate nature of seismic wave propagation, we can effectively analyze and interpret the distribution of valuable energy resources like petroleum and natural gas [2].…”

“…One popular network architecture for segmentation tasks is U-Net [8], which has been successfully applied in various domains, including cell segmentation. In the field of seismic data segmentation, researchers have extensively utilized U-Net to achieve remarkable results [2], [9], [10]. To address the challenge of limited labeled seismic data, Ferreira et al [11] incorporated generative adversarial networks (GANs) to generate synthetic seismic images.…”