Dataset Versions

v12

2024-11-16 7:45am

Generated on Nov 16, 2024

Dataset Split

Train Set 95%
774Images
Valid Set 4%
35Images
Test Set 1%
10Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 1024x1024

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -14° and +14°
Shear: ±12° Horizontal, ±12° Vertical
Saturation: Between -30% and +30%
Exposure: Between -14% and +14%
Blur: Up to 2.5px
Noise: Up to 1.05% of pixels
Cutout: 8 boxes with 5% size each
Mosaic: Applied