Simultaneous Automatic Picking and Manual Picking Refinement for First-Break

2025-02-03

IEEE Transactions on Geoscience and Remote Sensing

AI Summary

Manual first-arrival annotations in microseismic data often contain noise and outliers, degrading the performance of automated picking models. Method: This paper proposes the SPR framework, which—uniquely—models true first arrivals as latent variables and incorporates a prior over arrival annotations. Leveraging probabilistic graphical modeling and Bayesian inference, SPR jointly optimizes first-arrival prediction and manual label correction during deep neural network training, enabling robust co-modeling of both label noise and signal noise. The framework is architecture-agnostic and supports cross-site transfer learning. Contribution/Results: Evaluated on public datasets, SPR significantly improves picking accuracy, achieves tens-of-millisecond first-arrival offset correction, and demonstrates strong generalization capability and noise robustness.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Probabilistic Circuits and Graphical Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

First-break picking is a pivotal procedure in processing microseismic data for geophysics and resource exploration. Recent advancements in deep learning have catalyzed the evolution of automated methods for identifying first-break. Nevertheless, the complexity of seismic data acquisition and the requirement for detailed, expert-driven labeling often result in outliers and potential mislabeling within manually labeled datasets. These issues can negatively affect the training of neural networks, necessitating algorithms that handle outliers or mislabeled data effectively. We introduce the simultaneous picking and refinement (SPR) algorithm, designed to handle datasets plagued by outlier samples or even noisy labels. Unlike conventional approaches that regard manual picks as ground truth, our method treats the true first-break as a latent variable within a probabilistic model that includes a first-break labeling prior. SPR aims to uncover this variable, enabling dynamic adjustments and improved accuracy across the dataset. This strategy mitigates the impact of outliers or inaccuracies in manual labels. Intra-site picking experiments and cross-site generalization experiments on publicly available data confirm our method’s performance in identifying first-break and its generalization across different sites. Additionally, our investigations into noisy signals and labels underscore SPR’s resilience to both types of noise and its capability to refine misaligned manual annotations. Moreover, the flexibility of SPR, not being limited to any single network architecture, enhances its adaptability across various deep learning-based picking methods. Focusing on learning from data that may contain outliers or partial inaccuracies, SPR provides a robust solution to some of the principal obstacles in automatic first-break picking.

Problem

• Automatic P-wave Identification
• Error-prone Annotations
• Noisy Data Handling

Innovation

• Simultaneous Pickup and Repair (SPR) Algorithm
• Probability Model
• Automatic First-Arrival Wave Identification