Dataset Versions

v1

2024-10-17 9:21am

Generated on Oct 17, 2024

Dataset Split

Train Set 92%
339Images
Valid Set 5%
20Images
Test Set 3%
10Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Crop: 29% Minimum Zoom, 49% Maximum Zoom
Hue: Between -83° and +83°
Cutout: 7 boxes with 14% size each
Mosaic: Applied
Bounding Box: Crop: 14% Minimum Zoom, 34% Maximum Zoom
Bounding Box: Shear: ±14° Horizontal, ±17° Vertical
Bounding Box: Noise: Up to 0.81% of pixels

Similar Projects

See More