Dataset Versions
Versions
v1
2024-11-24 12:12am
Generated on Nov 23, 2024
3929 Total Images
View All Images
Dataset Split
Train Set 90%
3546Images
Valid Set 6%
238Images
Test Set 4%
145Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 26% Maximum Zoom
Rotation: Between -13° and +13°
Shear: ±15° Horizontal, ±15° Vertical
Grayscale: Apply to 18% of images
Hue: Between -14° and +14°
Saturation: Between -26% and +26%
Brightness: Between -24% and +24%
Exposure: Between -12% and +12%
Blur: Up to 3.4px
Noise: Up to 1.96% of pixels
Cutout: 5 boxes with 8% size each