Dataset Versions

v4

2022-12-15 11:26pm

Generated on Dec 15, 2022

Dataset Split

Train Set 79%
2838Images
Valid Set 11%
381Images
Test Set 11%
387Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Saturation: Between -26% and +26%
Blur: Up to 4.75px
Noise: Up to 5% of pixels