Dataset Versions

v9

2024-06-03 9:25am

Generated on Jun 3, 2024

Dataset Split

Train Set 92%
1101Images
Valid Set 6%
70Images
Test Set 2%
25Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Rotation: Between -15° and +15°
Grayscale: Apply to 20% of images
Brightness: Between -21% and +21%
Blur: Up to 1.5px
Noise: Up to 1.85% of pixels