Seghaj Dataset
v2
2023-02-03 2:20am
Generated on Feb 2, 2023
118 Total Images
View All Images
Dataset Split
Train Set 92%
108Images
Valid Set 6%
7Images
Test Set 3%
3Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -45° and +45°
Shear: ±23° Horizontal, ±23° Vertical
Brightness: Between -25% and +25%
Blur: Up to 3.25px
Noise: Up to 8% of pixels
Cutout: 7 boxes with 15% size each
Mosaic: Applied
Bounding Box: Flip: Horizontal, Vertical
Bounding Box: 90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Bounding Box: Crop: 0% Minimum Zoom, 15% Maximum Zoom
Bounding Box: Rotation: Between -35° and +35°
Bounding Box: Exposure: Between -25% and +25%