Dataset Versions

v1

2024-08-25 6:23pm

Generated on Aug 25, 2024

Dataset Split

Train Set 82%
45Images
Valid Set 9%
5Images
Test Set 9%
5Images

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 -1° and +1°
Hue: Between -57° and +57°
Saturation: Between -48% and +48%
Blur: Up to 1.7px
Noise: Up to 1.96% of pixels