Dataset Versions

v3

2024-07-15 3:41am

Generated on Jul 15, 2024

Dataset Split

Train Set 88%
2034Images
Valid Set 8%
192Images
Test Set 4%
97Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization

Augmentations

Outputs per training example: 3
Grayscale: Apply to 15% of images
Exposure: Between -10% and +10%
Blur: Up to 2.5px
Noise: Up to 0.27% of pixels