Dataset Versions

v8

June16-2024

Generated on Jun 17, 2024

Dataset Split

Train Set 88%
312Images
Valid Set 8%
29Images
Test Set 4%
15Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (black edges) in 1024x1024

Augmentations

Outputs per training example: 3
Hue: Between -15° and +15°
Saturation: Between -25% and +25%
Exposure: Between -10% and +10%
Blur: Up to 2.2px
Noise: Up to 1.84% of pixels