Dataset Versions

v1

2024-03-22 2:48pm

Generated on Mar 22, 2024

Dataset Split

Train Set 92%
17930Images
Valid Set 5%
1025Images
Test Set 3%
512Images

Preprocessing

Auto-Orient: Applied
Static Crop: 25-75% Horizontal Region, 25-75% Vertical Region
Resize: Stretch to 640x640
Filter Null: Require at least 3% of images to contain annotations.

Augmentations

Outputs per training example: 5
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -15° and +15°
Hue: Between -15° and +15°
Saturation: Between -25% and +25%
Exposure: Between -10% and +10%