README.md

---
tags:
- seismic
- ground-roll
- denoising
- seg-c3
- geophysics
- synthetic
task_categories:
- image-to-image
- other
size_categories:
- 10M-100M
pretty_name: SEG C3 Ground-Roll Dataset
viewer: false
---

# SEG C3 Ground-Roll Dataset

Paired noisy-input / noise-label SEG-Y volumes for supervised ground-roll attenuation, derived from the SEG C3 synthetic velocity model.

## Task

**Noise-label regression**: given a noisy pre-stack shot gather, predict the additive ground-roll noise component. The clean signal is recovered as:

```
denoised = noisy_input - predicted_noise
```

The noise labels serve as regression targets. Both input and label are 3D SEG-Y volumes with identical geometry.

## Dataset Description

- **Source**: SEG C3 synthetic velocity model ([wiki.seg.org/wiki/C3](https://wiki.seg.org/wiki/C3))
- **Geometry**: 9 regular shot gathers, 201 traces × 625 time samples per shot, dt = 2 ms
- **Noise modeling**: Reflection signals modeled with the acoustic wave equation; ground roll modeled with the elastic wave equation to capture its dispersive, low-velocity character
- **Noise intensity levels**: 1.0, 1.5, 2.0, 2.5, 3.0, 4.5, 5.0, 7.0, 9.0
- **Format**: Pre-stack SEG-Y (revision 1), IBM float
### Example 1

<div align="center">

  <p><b>Noisy Data</b></p>
  <img src="assets/ground_roll/noisy1.png" alt="Example 1 noisy data" width="92%">

  <p><b>Clean Data</b></p>
  <img src="assets/ground_roll/clean1.png" alt="Example 1 clean data" width="92%">

  <p><b>Ground-Roll Noise Label</b></p>
  <img src="assets/ground_roll/noise1.png" alt="Example 1 ground-roll noise label" width="92%">

</div>

<div align="center"><i>representative ground-roll attenuation sample at noise level 3.0.</i></div>

### Example 2

<div align="center">

  <p><b>Noisy Data</b></p>
  <img src="assets/ground_roll/noisy2.png" alt="Example 2 noisy data" width="92%">

  <p><b>Clean Data</b></p>
  <img src="assets/ground_roll/clean2.png" alt="Example 2 clean data" width="92%">

  <p><b>Ground-Roll Noise Label</b></p>
  <img src="assets/ground_roll/noise2.png" alt="Example 2 ground-roll noise label" width="92%">

</div>

<div align="center"><i>representative ground-roll attenuation sample at noise level 3.0.</i></div>

### Example 3

<div align="center">

  <p><b>Noisy Data</b></p>
  <img src="assets/ground_roll/noisy3.png" alt="Example 3 noisy data" width="92%">

  <p><b>Clean Data</b></p>
  <img src="assets/ground_roll/clean3.png" alt="Example 3 clean data" width="92%">

  <p><b>Ground-Roll Noise Label</b></p>
  <img src="assets/ground_roll/noise3.png" alt="Example 3 ground-roll noise label" width="92%">

</div>

<div align="center"><i>representative ground-roll attenuation sample at noise level 3.0.</i></div>


## File Structure

Each noise level has a matched pair of SEG-Y files:

| Level | Noisy Input | Noise Label | Size (approx) |
|-------|------------|-------------|---------------|
| 1.0 | SEGC3_shots1_9_noisy_1.0.sgy | SEGC3_shots1_9_noise_1.0.sgy | ~951 MB × 2 |
| 1.5 | SEGC3_shots1_9_noisy_1.5.sgy | SEGC3_shots1_9_noise_1.5.sgy | ~951 MB × 2 |
| 2.0 | SEGC3_shots1_9_noisy_2.0.sgy | SEGC3_shots1_9_noise_2.0.sgy | ~951 MB × 2 |
| 2.5 | SEGC3_shots1_9_noisy_2.5.sgy | SEGC3_shots1_9_noise_2.5.sgy | ~951 MB × 2 |
| 3.0 | SEGC3_shots1_9_noisy_3.0.sgy | SEGC3_shots1_9_noise_3.0.sgy | ~951 MB × 2 |
| 4.5 | SEGC3_shots1_9_noisy_4.5.sgy | SEGC3_shots1_9_noise_4.5.sgy | ~951 MB × 2 |
| 5.0 | SEGC3_shots1_9_noisy_5.0.sgy | SEGC3_shots1_9_noise_5.0.sgy | ~951 MB × 2 |
| 7.0 | SEGC3_shots1_9_noisy_7.0.sgy | SEGC3_shots1_9_noise_7.0.sgy | ~951 MB × 2 |
| 9.0 | SEGC3_shots1_9_noisy_9.0.sgy | SEGC3_shots1_9_noise_9.0.sgy | ~951 MB × 2 |

**Total**: 9 noisy + 9 noise SEG-Y files

## Loading Data

```python
import segyio
import numpy as np

def read_shot_gather(path, traces_per_shot=201):
    '''Read a regular SEG-Y file into (n_shots, n_traces, n_time).'''
    with segyio.open(path, "r", strict=False) as src:
        n_traces_total = src.tracecount
        n_shots = n_traces_total // traces_per_shot
        n_time = src.samples.size
        data = np.zeros((n_shots, traces_per_shot, n_time), dtype=np.float32)
        for i in range(n_shots):
            for j in range(traces_per_shot):
                data[i, j, :] = src.trace[i * traces_per_shot + j]
    return data

# Load a level-3.0 pair
noisy  = read_shot_gather("noisy/SEGC3_shots1_9_noisy_3.0.sgy")
noise  = read_shot_gather("noise/SEGC3_shots1_9_noise_3.0.sgy")
signal = noisy - noise   # clean reference
```

With `huggingface_hub`:

```python
from huggingface_hub import hf_hub_download

path = hf_hub_download(
    repo_id="GeoBrain/ground-roll",
    filename="noisy/SEGC3_shots1_9_noisy_3.0.sgy",
    repo_type="dataset",
)
```

## Train / Val / Test Split

Shot-level sequential split by FFID (field file ID), avoiding trace leakage:

| Split | Shots | Fraction |
|-------|-------|----------|
| Train | 7     | 77.8%    |
| Val   | 1     | 11.1%    |
| Test  | 1     | 11.1%    |

The split is done at loading time (not pre-saved as separate files) so users can adjust the ratios.

## Preprocessing Recipe

The companion benchmark applies:

1. **Normalization**: `max_abs`, global scope — the entire noisy volume scaled to [-1, 1]; same stats applied to the noise label
2. **Patching**: Overlapping 2D patches (128 traces × 256 time samples), 50% overlap, yielding (1, H, W) tensors

No spherical-divergence correction is applied (raw amplitudes are used).

## Benchmark Results

See the companion model repository for full benchmark results across UNet, ResUNet, DnCNN, and Attention UNet architectures at each noise level.

## Citation

If you use this dataset, please cite the SEG C3 model and the companion benchmark:

```bibtex
@misc{seg_c3_ground_roll,
  title={SEG C3 Ground-Roll Attenuation Benchmark},
  howpublished={https://huggingface.co/datasets/GeoBrain/ground-roll},
}
```

## References

- SEG C3 Velocity Model: https://wiki.seg.org/wiki/C3
- `segyio` library: https://github.com/equinor/segyio