Dataset Versions

v3

2024-04-19 11:10pm

Generated on Apr 19, 2024

Dataset Split

Train Set 82%
960Images
Valid Set 10%
116Images
Test Set 8%
88Images

Preprocessing

Auto-Orient: Applied
Static Crop: 23-83% Horizontal Region, 15-66% Vertical Region
Resize: Fit (reflect edges) in 640x640
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Blur: Up to 1.7px
Noise: Up to 1.09% of pixels