Dataset Versions

v2

2024-06-07 2:04pm

Generated on Jun 7, 2024

Dataset Split

Train Set 88%
408Images
Valid Set 8%
39Images
Test Set 4%
19Images

Preprocessing

Auto-Orient: Applied
Static Crop: 12-90% Horizontal Region, 32-88% Vertical Region
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization

Augmentations

Outputs per training example: 3
Flip: Vertical
Rotation: Between -16° and +16°
Shear: ±13° Horizontal, ±8° Vertical