Dataset Versions

v1

2024-04-13 4:06pm

Generated on Apr 13, 2024

Dataset Split

Train Set 88%
630Images
Valid Set 8%
60Images
Test Set 4%
30Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Grayscale: Apply to 19% of images
Hue: Between -15° and +15°
Saturation: Between -29% and +29%
Brightness: Between -26% and +26%
Exposure: Between -25% and +25%
Blur: Up to 1.6px
Noise: Up to 1.8% of pixels
Cutout: 5 boxes with 10% size each