Dataset Versions

v3

2024-07-30 4:31am

Generated on Jul 30, 2024

Dataset Split

Train Set 94%
417Images
Valid Set 4%
18Images
Test Set 2%
10Images

Preprocessing

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

Augmentations

Outputs per training example: 3
Flip: Horizontal
Rotation: Between -26° and +26°
Exposure: Between -15% and +15%
Blur: Up to 1.2px
Noise: Up to 1.33% of pixels