Dataset Versions

v2

2024-10-19 10:47am

Generated on Oct 19, 2024

Dataset Split

Train Set 88%
42Images
Valid Set 8%
4Images
Test Set 4%
2Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Rotation: Between -21° and +21°
Shear: ±14° Horizontal, ±19° Vertical
Saturation: Between -34% and +34%
Bounding Box: Exposure: Between -10% and +10%
Bounding Box: Blur: Up to 2.5px
Bounding Box: Noise: Up to 1.64% of pixels