Dataset Versions

v2

v2

Generated on Jul 7, 2024

Dataset Split

Train Set 90%
843Images
Valid Set 7%
62Images
Test Set 4%
34Images

Preprocessing

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

Augmentations

Outputs per training example: 3
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Shear: ±10° Horizontal, ±10° Vertical
Grayscale: Apply to 15% of images