Dataset Versions

v5

test4npass

Generated on Oct 24, 2023

Dataset Split

Train Set 88%
159Images
Valid Set 8%
15Images
Test Set 4%
7Images

Preprocessing

Auto-Orient: Applied
Resize: Fit within 640x640
Auto-Adjust Contrast: Using Contrast Stretching

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Grayscale: Apply to 40% of images
Brightness: Between -50% and +50%
Blur: Up to 4px
Bounding Box: Flip: Horizontal, Vertical