What question did this study set out to answer?

The aim is to evaluate the performance of deep learning models for accurate seismic fault detection.

June 3, 2026Open Access

Improving seismic fault detection through fault-balanced patch extraction and deep learning networks (UNet, Efficient-UNet, and VGG19-UNet)

KNKeyvan NajafzadehUniversity of Tehran MNMohammad Emami NiriPetroleum University of Technology ABAbbas BahroudiUniversity of Tehran

Key Points

The aim is to evaluate the performance of deep learning models for accurate seismic fault detection.
Three models (UNet, Efficient-UNet, VGG19-UNet) were trained on 2D seismic patches from a real 3D volume.
A fault-aware dataset preparation strategy was utilized to address class imbalance, including data augmentation.
Models were evaluated using Dice and IoU metrics on 128 × 128 patches for performance assessment.
Standard UNet achieved the highest overall performance with mean Dice of 0.91 and IoU of 0.85.
VGG19-UNet showed competitive performance with Dice of 0.88 and IoU of 0.81.
Efficient-UNet exhibited lower accuracy with Dice of 0.73 and IoU of 0.61, especially for thin faults.

Abstract

Accurate seismic fault detection is essential for reliable structural interpretation, reservoir characterization, and drilling risk mitigation. Conventional attribute-based methods often suffer from noise sensitivity, fragmented fault continuity, and strong dependence on interpreter experience, particularly in structurally complex settings. This study evaluates the performance of three convolutional neural network–based segmentation models, Standard UNet, Efficient-UNet, and VGG19-UNet, for automatic fault detection using 2D seismic patches extracted from a real 3D seismic volume. To address the severe class imbalance inherent in fault interpretation, a fault-aware dataset preparation strategy was implemented. Training samples were extracted from fault-rich regions using a minimum fault-pixel threshold, combined with controlled fault-mask dilation and data augmentation. All models were trained using a unified Dice + Binary Cross-Entropy loss function and evaluated on 128 × 128 patches using both overlap-based metrics (Dice and IoU) and distance-tolerant measures to assess spatial robustness. The results show that the Standard UNet achieves the highest overall performance, with a mean Dice of 0.91 and IoU of 0.85, demonstrating strong balance between precision and recall. The VGG19-UNet provides competitive performance (Dice = 0.88, IoU = 0.81) with relatively consistent fault continuity, while the Efficient-UNet exhibits lower accuracy (Dice = 0.73, IoU = 0.61), particularly in detecting thin or low-contrast faults. Distance-tolerant evaluation further confirms the robustness of the Standard UNet, with F1 scores exceeding 0.98 at small spatial tolerances. These results highlight the importance of both dataset design and architectural choice in deep learning–based seismic fault detection. The proposed framework demonstrates strong performance on the studied dataset, while its broader applicability to other geological settings remains an important direction for future work.

Ask AI

Helpful

Bookmark

View Full Paper