Dataset Versions
v3
2024-02-08 7:28pm
Generated on Feb 8, 2024
6407 Total Images
View All Images
Dataset Split
Train Set 92%
5865Images
Valid Set 6%
361Images
Test Set 3%
181Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Filter Null: Require all images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -14° and +14°
Shear: ±9° Horizontal, ±11° Vertical
Saturation: Between -24% and +24%
Brightness: Between -20% and +20%
Blur: Up to 2.7px
Noise: Up to 0.69% of pixels