Dataset Versions
Versions
v1
2023-03-19 10:01pm
Generated on Mar 19, 2023
110 Total Images
View All Images
Dataset Split
Train Set 98%
108Images
Valid Set 1%
1Images
Test Set 1%
1Images
Preprocessing
Auto-Orient: Applied
Resize: Fit within 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Crop: 0% Minimum Zoom, 1% Maximum Zoom
Rotation: Between -5° and +5°
Shear: ±12° Horizontal, ±9° Vertical
Saturation: Between -60% and +60%
Brightness: Between -45% and +45%
Exposure: Between -7% and +7%
Blur: Up to 2px
Noise: Up to 1% of pixels