Dataset Versions
v4
2024-09-09 6:49am
Generated on Sep 9, 2024
1426 Total Images
View All Images
Dataset Split
Train Set 88%
1248Images
Valid Set 8%
118Images
Test Set 4%
60Images
Preprocessing
Auto-Orient: Applied
Static Crop: 11-92% Horizontal Region, 17-92% Vertical Region
Resize: Fit within 640x320
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -6° and +6°
Shear: ±6° Horizontal, ±6° Vertical
Hue: Between -12° and +12°
Saturation: Between -15% and +15%
Brightness: Between -20% and +20%
Exposure: Between -10% and +10%
Blur: Up to 0.5px
Noise: Up to 0.18% of pixels