Dataset Versions

v4

2024-05-17 2:07pm

Generated on May 17, 2024

Dataset Split

Train Set 82%
3532Images
Valid Set 12%
504Images
Test Set 6%
253Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization
Grayscale: Applied

Augmentations

Outputs per training example: 2
Hue: Between -15° and +15°
Blur: Up to 2.5px
Noise: Up to 0.1% of pixels