Skip to content

Dataset · blanchon

AID aerial images of 30 landscape and object types

blanchon/AID

10,000 Google Earth RGB satellite images for bird's-eye scene classification.

What's inside

The dataset includes 10,000 color RGB images of 600x600 pixels extracted from Google Earth across different countries and regions worldwide. Images are distributed into 30 scene categories (such as airports and other terrain types) with several hundred examples per class. Metadata and class names are in English.

What you can do with it

  • Training and fine-tuning image classification models on aerial photography data
  • Testing neural networks for satellite map analysis and territorial monitoring
Tasks
Image classification
Languages
English
Size
10K–100K rows
Downloads this month
22.5K
Likes
9

How to load it

from datasets import load_dataset

ds = load_dataset("blanchon/AID")