Dataset Versions

v8

2024-02-16 4:15pm

Generated on Feb 16, 2024

Dataset Split

Train Set 85%
1368Images
Valid Set 9%
152Images
Test Set 5%
84Images

Preprocessing

Auto-Orient: Applied
Resize: Fit within 720x720
Auto-Adjust Contrast: Using Contrast Stretching

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 30% Maximum Zoom
Rotation: Between -6° and +6°
Shear: ±5° Horizontal, ±5° Vertical
Hue: Between -15° and +15°
Saturation: Between -30% and +30%
Brightness: Between -18% and +18%
Exposure: Between -15% and +15%
Blur: Up to 3px
Noise: Up to 0.57% of pixels

Similar Projects

See More