Dataset Versions
v7
2023-01-17 10:06am
Generated on Jan 17, 2023
1651 Total Images
View All Images
Dataset Split
Train Set 89%
1467Images
Valid Set 7%
122Images
Test Set 4%
62Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 65% Maximum Zoom
Rotation: Between -14° and +14°
Shear: ±22° Horizontal, ±34° Vertical
Grayscale: Apply to 40% of images
Hue: Between -76° and +76°
Brightness: Between -38% and +38%
Exposure: Between -34% and +34%
Blur: Up to 5.25px
Noise: Up to 7% of pixels
Cutout: 10 boxes with 5% size each
Mosaic: Applied
Bounding Box: 90° Rotate: Clockwise, Counter-Clockwise