Dataset Versions

v1

2024-05-26 12:33am

Generated on May 25, 2024

Dataset Split

Train Set 86%
507Images
Valid Set 7%
40Images
Test Set 7%
40Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x840
Grayscale: Applied

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Brightness: Between -25% and +25%
Exposure: Between -15% and +15%
Noise: Up to 0.14% of pixels
Bounding Box: Crop: 0% Minimum Zoom, 20% Maximum Zoom
Bounding Box: Rotation: Between -15° and +15°
Bounding Box: Shear: ±10° Horizontal, ±10° Vertical
Bounding Box: Blur: Up to 2.5px