Dataset Versions
v6
2023-08-01 12:35pm
Generated on Aug 1, 2023
385 Total Images
View All Images
Dataset Split
Train Set 81%
312Images
Valid Set 12%
47Images
Test Set 7%
26Images
Preprocessing
Auto-Orient: Applied
Isolate Objects: Applied
Resize: Fit (black edges) in 600x600
Auto-Adjust Contrast: Using Contrast Stretching
Grayscale: Applied
Modify Classes: 0 remapped, 1 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 15% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±9° Horizontal, ±8° Vertical
Hue: Between -13° and +13°
Brightness: Between -20% and +20%
Exposure: Between -8% and +8%
Blur: Up to 0.25px
Noise: Up to 3% of pixels