DeepStratNet: A Context-Aware Coordinate Regression Framework for Seismic Horizon Tracking under Sparse Labels

2026-10-01

AI Summary

This study addresses the challenges of seismic horizon tracking under sparse labels, where traditional segmentation methods suffer from substantial preprocessing errors, convergence difficulties, and a lack of inter-slice context. To overcome these limitations, this work pioneers a coordinate regression paradigm as an alternative to dense semantic segmentation. Specifically, horizon tracking is reformulated as bounded coordinate regression, directly predicting depth coordinates at each lateral position via a pretrained visual backbone coupled with LSTM-based sequence modeling. Geological priors are further incorporated as regularization constraints to enforce lateral continuity. Experimental results demonstrate that the proposed method outperforms segmentation baselines in both RMSE and PCC metrics while significantly reducing reliance on dense annotations. Moreover, prediction variance serves as an automated quality control indicator reflecting geological complexity.

Abstract

Automatic horizon tracking is a foundational task in 3D seismic interpretation. Most existing deep learning approaches formulate it as dense semantic segmentation, typically using U-Net-based architectures. The model produces a probability map over all pixels that must be post-processed to extract precise horizon coordinates, while horizon picks in time/depth must be converted into dense masks for training. Unpicked seismic traces are consequently treated as background, which can hinder convergence, and both pre- and post-processing can introduce errors into the final interpretation. Moreover, 2D segmentation models do not inherently capture inter-slice context, while 3D models are often computationally prohibitive. We instead formulate horizon tracking as a bounded coordinate regression problem, where the model directly predicts the time/depth coordinate of the target horizon at each lateral position. We propose a lightweight regression head compatible with any pretrained vision backbone, coupled with an LSTM module to model inter-slice context and produce a continuous horizon surface across the volume. A combination of L1 and L2 losses supervises predictions at valid horizon picks, while a geology-informed regularization enforces lateral continuity between successive traces. Under controlled experimental conditions, we evaluate four pretrained vision backbones under both segmentation and regression configurations on a seismic volume from New Zealand. The proposed approach consistently outperforms its segmentation counterparts quantitatively, using metrics including RMSE and PCC, and qualitatively, while also demonstrating greater robustness to increasing sparsity of training picks. Finally, we show that prediction variation across successive traces captures local variations in geological complexity, providing an automated quality control measure for downstream seismic interpretation.

Problem

• seismic horizon tracking
• sparse labels
• coordinate regression
• semantic segmentation
• 3D seismic interpretation

Innovation

• coordinate regression
• context-aware LSTM
• sparse labels
• geology-informed regularization
• seismic horizon tracking