Dataset Versions

v1

2024-11-28 7:32pm

Generated on Nov 28, 2024

Dataset Split

Train Set 87%
282Images
Valid Set 10%
33Images
Test Set 3%
9Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Contrast Stretching

Augmentations

Outputs per training example: 3
Flip: Horizontal
90° Rotate: Clockwise, Counter-Clockwise
Hue: Between -15° and +15°
Saturation: Between -25% and +25%
Exposure: Between -7% and +7%
Noise: Up to 0.1% of pixels