Dataset Versions

v1

2024-03-27 10:47am

Generated on Mar 27, 2024

Dataset Split

Train Set 92%
480Images
Valid Set 4%
20Images
Test Set 4%
20Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Grayscale: Apply to 24% of images
Hue: Between -24° and +24°
Saturation: Between -34% and +34%
Exposure: Between -15% and +15%
Blur: Up to 4.8px
Noise: Up to 1.99% of pixels