Dataset Versions

v9

2024-11-23 11:44pm

Generated on Nov 24, 2024

Dataset Split

Train Set 87%
5028Images
Valid Set 8%
480Images
Test Set 4%
244Images

Preprocessing

Auto-Orient: Applied
Resize: Fit within 416x416
Grayscale: Applied
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
Grayscale: Apply to 15% of images
Exposure: Between -10% and +10%
Noise: Up to 1.92% of pixels
Bounding Box: Shear: ±14° Horizontal, ±14° Vertical
Bounding Box: Brightness: Between -15% and +15%