Dataset Versions
v2
2023-10-17 11:58pm
Generated on Oct 18, 2023
1622 Total Images
View All Images
Dataset Split
Train Set 87%
1419Images
Valid Set 8%
135Images
Test Set 4%
68Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Filter Null: Require at least 48% of images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Saturation: Between -31% and +31%
Noise: Up to 2% of pixels