Dataset Versions

v7

2024-11-04 12:23pm

Generated on Nov 4, 2024

Dataset Split

Train Set 87%
552Images
Valid Set 9%
56Images
Test Set 4%
24Images

Preprocessing

Auto-Orient: Applied
Static Crop: 8-93% Horizontal Region, 17-75% Vertical Region
Auto-Adjust Contrast: Using Histogram Equalization
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Brightness: Between -15% and +15%
Exposure: Between -10% and +10%