Dataset Versions

v5

2024-10-18 8:19am

Generated on Oct 18, 2024

Dataset Split

Train Set 88%
606Images
Valid Set 8%
58Images
Test Set 4%
28Images

Preprocessing

Auto-Orient: Applied

Augmentations

Outputs per training example: 3
Flip: Vertical
Shear: ±10° Horizontal, ±10° Vertical
Hue: Between -65° and +65°
Noise: Up to 0.1% of pixels
Mosaic: Applied
Bounding Box: Exposure: Between -60% and +60%
Bounding Box: Blur: Up to 2.5px