Dataset Versions

v15

2024-04-29 9:40am

Generated on Apr 29, 2024

Dataset Split

Train Set 88%
588Images
Valid Set 8%
56Images
Test Set 4%
28Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (reflect edges) in 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -12° and +12°
Shear: ±12° Horizontal, ±12° Vertical
Grayscale: Apply to 15% of images
Hue: Between -20° and +20°