Dataset Versions

v1

2024-04-06 6:13pm

Generated on Apr 6, 2024

Dataset Split

Train Set 88%
2727Images
Valid Set 8%
261Images
Test Set 4%
126Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization
Grayscale: Applied
Tile: 3 rows x 3 columns

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -25° and +25°
Brightness: Between -38% and +38%
Blur: Up to 3px
Noise: Up to 6.04% of pixels
Cutout: 11 boxes with 10% size each
Bounding Box: Flip: Horizontal
Bounding Box: Rotation: Between -15° and +15°
Bounding Box: Shear: ±13° Horizontal, ±13° Vertical