---
license: mit
pretty_name: CarNet Crack Detection (Sun520, Rain365, BJN260)
task_categories:
- image-segmentation
tags:
- crack-detection
- pavement
- road
- semantic-segmentation
size_categories:
- 1K<n<10K
---

# crack

## Datasets

| Folder | Images | Scenario |
|--------|--------|----------|
| `BJN260_full_resolution` | 260 | Night scenes (Beijing) |
| `Rain365_full_resolution` | 365 | Post-rain daytime |
| `Sun520_full_resolution` | 520 | Sunny daytime |

Each dataset contains:

- `img/` — crack photos
- `gt/` — pixel-level ground-truth masks
- `Composite image/` — crack + mask overlays

## Citation

```
@article{li2021fast,
    title={Fast and Accurate Road Crack Detection Based on Adaptive Cost-Sensitive Loss Function},
    author={Kai Li and Bo Wang and Yingjie Tian and Zhiquan Qi},
    journal={IEEE Transactions on Cybernetics},
    pages={1-12},
    year={2021},
    doi={10.1109/TCYB.2021.3103885}
}
```