---
license: cc0-1.0
language:
- en
size_categories:
- n<1K
---

# Dubai Aerial Semantic Segmentation Dataset

## About the Dataset

Semantic segmentation of aerial imagery over Dubai, United Arab Emirates (UAE), into six land cover classes.

## Context

**Humans in the Loop** is publishing this open-access dataset, created in collaboration with the **Mohammed Bin Rashid Space Center (MBRSC)** in Dubai, UAE.

The dataset aims to support machine learning and computer vision tasks in urban analysis, environmental monitoring, and remote sensing.

## Dataset Description

- **Source**: Aerial imagery obtained from MBRSC satellites
- **Annotations**: Pixel-wise semantic segmentation
- **Total images**: 72
- **Grouping**: Images are organized into 6 larger tiles
- **Classes**: 6 semantic categories

### Semantic Classes and Color Codes

| Class               | Hex Color Code |
|---------------------|----------------|
| Building            | `#3C1098`       |
| Land (Unpaved Area) | `#8429F6`       |
| Road                | `#6EC1E4`       |
| Vegetation          | `#FEDD3A`       |
| Water               | `#E2A929`       |
| Unlabeled           | `#9B9B9B`       |

## Acknowledgements

The images were segmented by trainees of the **Roia Foundation** in Syria under the supervision of **Humans in the Loop**.

## License

This dataset is licensed under **[CC0 1.0 Universal (Public Domain Dedication)](https://creativecommons.org/publicdomain/zero/1.0/)** — free to use for any purpose.