Hibou Computer Vision: 10K-100K Images for YOLO Drone Detection

Hibou Computer Vision contains between 10,000 and 100,000 images for drone detection and localization, developed by the Hibou-Foundation. The data is specifically structured for training YOLO-based models and includes binary classification labels for drones and other objects. It was last updated in March 2026.

Use Cases

• Training YOLO-based object detection models using the image column
• Real-time drone localization benchmarking for the Drone class
• Feature extraction for aerial surveillance using optimized Parquet files

Strengths

• Scale of 10,000 to 100,000 records
• Specific class distribution provided (74.17% Drone)
• Optimized Parquet format for high-performance data loading

Limitations

• Class imbalance with drones representing nearly 75% of the data
• Limited to only two object classes

Provenance

Source

Hibou-Foundation

Freshness

Last updated March 2026

Geography

United States