Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Seismic First Break Picking in a Higher Dimension Using Deep Graph Learning

Hongtao Wang, Li Long, Jiangshe Zhang, Xiaoli Wei, Chunxia Zhang, Zhenbo Guo

TL;DR

This paper tackles first-break picking in seismic data by exploiting higher-dimensional context through a graph-based representation. It introduces DGL-FB, which builds a large graph with traces as nodes and similarity-based edges, applies subgraph sampling for scalability, encodes global subgraph information with a GraphSAGE-LSTM encoder, and fuses it with local trace signals in a 1D ResUNet for FB segmentation. The approach is evaluated on open-field datasets, showing improved accuracy (81.8% vs 76.3%) and a dramatic RMSE reduction (3.24 vs 460.0), demonstrating robust and stable FB detection. The work highlights the value of graph-based global information in seismic FB picking and suggests a viable path for higher-dimensional statics corrections in real surveys.

Abstract

Contemporary automatic first break (FB) picking methods typically analyze 1D signals, 2D source gathers, or 3D source-receiver gathers. Utilizing higher-dimensional data, such as 2D or 3D, incorporates global features, improving the stability of local picking. Despite the benefits, high-dimensional data requires structured input and increases computational demands. Addressing this, we propose a novel approach using deep graph learning called DGL-FB, constructing a large graph to efficiently extract information. In this graph, each seismic trace is represented as a node, connected by edges that reflect similarities. To manage the size of the graph, we develop a subgraph sampling technique to streamline model training and inference. Our proposed framework, DGL-FB, leverages deep graph learning for FB picking. It encodes subgraphs into global features using a deep graph encoder. Subsequently, the encoded global features are combined with local node signals and fed into a ResUNet-based 1D segmentation network for FB detection. Field survey evaluations of DGL-FB show superior accuracy and stability compared to a 2D U-Net-based benchmark method.

Seismic First Break Picking in a Higher Dimension Using Deep Graph Learning

TL;DR

This paper tackles first-break picking in seismic data by exploiting higher-dimensional context through a graph-based representation. It introduces DGL-FB, which builds a large graph with traces as nodes and similarity-based edges, applies subgraph sampling for scalability, encodes global subgraph information with a GraphSAGE-LSTM encoder, and fuses it with local trace signals in a 1D ResUNet for FB segmentation. The approach is evaluated on open-field datasets, showing improved accuracy (81.8% vs 76.3%) and a dramatic RMSE reduction (3.24 vs 460.0), demonstrating robust and stable FB detection. The work highlights the value of graph-based global information in seismic FB picking and suggests a viable path for higher-dimensional statics corrections in real surveys.

Abstract

Contemporary automatic first break (FB) picking methods typically analyze 1D signals, 2D source gathers, or 3D source-receiver gathers. Utilizing higher-dimensional data, such as 2D or 3D, incorporates global features, improving the stability of local picking. Despite the benefits, high-dimensional data requires structured input and increases computational demands. Addressing this, we propose a novel approach using deep graph learning called DGL-FB, constructing a large graph to efficiently extract information. In this graph, each seismic trace is represented as a node, connected by edges that reflect similarities. To manage the size of the graph, we develop a subgraph sampling technique to streamline model training and inference. Our proposed framework, DGL-FB, leverages deep graph learning for FB picking. It encodes subgraphs into global features using a deep graph encoder. Subsequently, the encoded global features are combined with local node signals and fed into a ResUNet-based 1D segmentation network for FB detection. Field survey evaluations of DGL-FB show superior accuracy and stability compared to a 2D U-Net-based benchmark method.
Paper Structure (6 sections, 5 equations, 3 figures)

This paper contains 6 sections, 5 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Method
  3. Graph Data Preparation
  4. DGL-FB
  5. Experimental Results
  6. Conclusion

Figures (3)

  • Figure 1: A showcase of the FB picking task. The left subfigure shows the FB of a single trace. The right subfigure indicates a high correlation among the FBs of the adjacent traces. The red dotted box indicates the location of the single-trace signal in the left subfigure.
  • Figure 2: ResUNet-1D Network Structure
  • Figure 3: Two picking cases of a common shot gather from the test survey Lalor using DGL-FB.