Dataset Versions

v6

2024-08-28 10:43pm

Generated on Aug 28, 2024

Dataset Split

Train Set 91%
753Images
Valid Set 7%
57Images
Test Set 2%
20Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -25° and +25°
Shear: ±32° Horizontal, ±14° Vertical
Blur: Up to 1.3px
Cutout: 4 boxes with 14% size each