Dataset Versions
v3
2023-11-01 10:03pm
Generated on Nov 1, 2023
145 Total Images
View All Images
Dataset Split
Train Set 85%
123Images
Valid Set 9%
13Images
Test Set 6%
9Images
Preprocessing
Auto-Orient: Applied
Static Crop: 8-75% Horizontal Region, 19-64% Vertical Region
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -29° and +29°
Grayscale: Apply to 45% of images
Saturation: Between -45% and +45%
Brightness: Between -28% and +28%
Blur: Up to 2.5px
Bounding Box: Shear: ±4° Horizontal, ±15° Vertical
Bounding Box: Noise: Up to 9% of pixels