Dataset Versions

v1

2024-12-16 9:54pm

Generated on Dec 16, 2024

Dataset Split

Train Set 90%
1155Images
Valid Set 7%
87Images
Test Set 3%
45Images

Preprocessing

Auto-Orient: Applied
Static Crop: 16-90% Horizontal Region, 17-83% Vertical Region
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 16% Maximum Zoom
Rotation: Between -11° and +11°
Shear: ±9° Horizontal, ±6° Vertical
Saturation: Between -18% and +18%
Brightness: Between -12% and +12%
Exposure: Between -6% and +6%
Noise: Up to 0.26% of pixels