Dataset Versions

v3

2023-07-03 6:01pm

Generated on Jul 3, 2023

Dataset Split

Train Set 87%
498Images
Valid Set 8%
47Images
Test Set 4%
25Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Grayscale: Applied

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Rotation: Between -45° and +45°
Shear: ±19° Horizontal, ±14° Vertical
Hue: Between -25° and +25°
Bounding Box: Flip: Horizontal
Bounding Box: Rotation: Between -15° and +15°
Bounding Box: Brightness: Between -45% and +45%
Bounding Box: Noise: Up to 5% of pixels