Dataset Versions

v2

2023-10-25 9:06pm

Generated on Oct 26, 2023

Dataset Split

Train Set 88%
282Images
Valid Set 8%
24Images
Test Set 4%
13Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization

Augmentations

Outputs per training example: 3
Rotation: Between -15° and +15°
Shear: ±15° Horizontal, ±15° Vertical
Brightness: Between -25% and +25%