Dataset Versions

v2

2024-09-22 2:45pm

Generated on Sep 22, 2024

Dataset Split

Train Set 87%
1179Images
Valid Set 8%
113Images
Test Set 4%
56Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Rotation: Between -15° and +15°
Shear: ±10° Horizontal, ±10° Vertical
Grayscale: Apply to 15% of images
Brightness: Between -15% and +15%
Exposure: Between -10% and +10%
Blur: Up to 2.5px
Noise: Up to 0.1% of pixels