Dataset Versions

v2

2023-12-25 9:39am

Generated on Dec 25, 2023

Dataset Split

Train Set 88%
1251Images
Valid Set 8%
118Images
Test Set 4%
59Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Vertical
Crop: 0% Minimum Zoom, 12% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±10° Horizontal, ±10° Vertical
Brightness: Between -25% and +25%
Blur: Up to 0.5px
Noise: Up to 3% of pixels
Cutout: 4 boxes with 7% size each