Dataset Versions

v4

2024-10-17 12:20pm

Generated on Oct 17, 2024

Dataset Split

Train Set 95%
2928Images
Valid Set 3%
108Images
Test Set 2%
53Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Vertical
Rotation: Between -15° and +15°
Shear: ±12° Horizontal, ±14° Vertical
Saturation: Between -28% and +28%
Brightness: Between -22% and +22%
Noise: Up to 1.01% of pixels