Dataset Versions

v4

2024-11-05 6:43pm

Generated on Nov 5, 2024

Dataset Split

Train Set 92%
36Images
Valid Set 3%
1Images
Test Set 5%
2Images

Preprocessing

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

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Blur: Up to 2.5px
Noise: Up to 0.97% of pixels