Dataset Versions

v6

f_v

Generated on Feb 2, 2024

Dataset Split

Train Set 89%
1626Images
Valid Set 7%
130Images
Test Set 4%
73Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 18% Maximum Zoom
Rotation: Between -10° and +10°
Shear: ±12° Horizontal, ±8° Vertical
Saturation: Between -33% and +33%
Brightness: Between -9% and +9%
Exposure: Between -12% and +12%
Noise: Up to 1.06% of pixels