Dataset Versions

v2

2024-01-17 1:04pm

Generated on Jan 17, 2024

Dataset Split

Train Set 74%
3978Images
Valid Set 20%
1097Images
Test Set 5%
292Images

Preprocessing

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

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -15° and +15°
Saturation: Between -34% and +34%
Brightness: Between -15% and +15%
Noise: Up to 0.65% of pixels