Dataset Versions

v4

2024-07-02 7:21am

Generated on Jul 2, 2024

Dataset Split

Train Set 99%
7356Images
Valid Set %
0Images
Test Set 1%
100Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (black edges) in 1024x768
Modify Classes: 1 remapped, 0 dropped
Filter Null: Require at least 95% of images to contain annotations.

Augmentations

Outputs per training example: 2
Flip: Horizontal
Shear: ±7° Horizontal, ±7° Vertical
Exposure: Between -9% and +9%
Noise: Up to 1.76% of pixels
Cutout: 5 boxes with 5% size each