Dataset Versions

v1

2024-05-08 12:52pm

Generated on May 8, 2024

Dataset Split

Train Set 88%
3432Images
Valid Set 8%
327Images
Test Set 4%
163Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 128x72

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -15° and +15°
Shear: ±10° Horizontal, ±10° Vertical
Brightness: Between -15% and +15%
Noise: Up to 1.09% of pixels