Dataset Versions
v5
2023-04-13 2:33pm
Generated on Apr 13, 2023
608 Total Images
View All Images
Dataset Split
Train Set 87%
531Images
Valid Set 8%
50Images
Test Set 4%
27Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Crop: 0% Minimum Zoom, 13% Maximum Zoom
Rotation: Between -9° and +9°
Shear: ±14° Horizontal, ±10° Vertical
Saturation: Between -6% and +6%
Brightness: Between -8% and +8%
Exposure: Between -15% and +15%
Blur: Up to 1.75px
Noise: Up to 2% of pixels