Dataset Versions

v1

2024-04-11 8:10am

Generated on Apr 11, 2024

Dataset Split

Train Set 96%
372Images
Valid Set %
0Images
Test Set 4%
14Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -3° and +3°
Saturation: Between -5% and +5%
Brightness: Between -8% and +8%
Blur: Up to 0.7px
Noise: Up to 0.1% of pixels