Dataset Versions
v2
2023-09-19 9:56am
Generated on Sep 19, 2023
3597 Total Images
View All Images
Dataset Split
Train Set 87%
3147Images
Valid Set 8%
300Images
Test Set 4%
150Images
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, 64% Maximum Zoom
Rotation: Between -21° and +21°
Grayscale: Apply to 10% of images
Saturation: Between -78% and +78%
Brightness: Between -51% and +51%
Exposure: Between -25% and +25%
Blur: Up to 2.5px
Noise: Up to 9% of pixels
Cutout: 6 boxes with 7% size each