Dataset Versions

v15

2024-08-24 10:34am

Generated on Aug 24, 2024

Dataset Split

Train Set 88%
2523Images
Valid Set 10%
290Images
Test Set 2%
70Images

Preprocessing

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

Augmentations

Outputs per training example: 3
Flip: Vertical
Hue: Between -25° and +25°
Brightness: Between -25% and +25%
Blur: Up to 1.3px
Noise: Up to 1.99% of pixels