Dataset Versions
v1
V1 - Grayscale - High Rotation
Generated on Feb 27, 2024
597 Total Images
View All Images
Dataset Split
Train Set 87%
522Images
Valid Set 8%
50Images
Test Set 4%
25Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Contrast Stretching
Grayscale: Applied
Filter Null: Require at least 86% of images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 8% Maximum Zoom
Rotation: Between -22° and +22°
Shear: ±13° Horizontal, ±13° Vertical
Hue: Between -16° and +16°
Saturation: Between -20% and +20%
Brightness: Between -12% and +12%
Exposure: Between -13% and +13%
Blur: Up to 1.8px
Noise: Up to 0.65% of pixels