Dataset Versions

v7

2023-12-03 11:18am

Generated on Dec 3, 2023

Dataset Split

Train Set 88%
1512Images
Valid Set 8%
144Images
Test Set 4%
72Images

Preprocessing

Auto-Orient: Applied
Static Crop: 25-75% Horizontal Region, 25-75% Vertical Region
Resize: Stretch to 640x640
Filter Null: Require at least 40% of images to contain annotations.

Augmentations

Outputs per training example: 3
Rotation: Between -30° and +30°
Shear: ±10° Horizontal, ±10° Vertical
Exposure: Between -25% and +25%
Blur: Up to 2.75px