Dataset Versions
v4
2023-03-28 12:28pm
Generated on Mar 28, 2023
1584 Total Images
View All Images
Dataset Split
Train Set 91%
1440Images
Valid Set 4%
64Images
Test Set 5%
80Images
Preprocessing
Auto-Orient: Applied
Auto-Adjust Contrast: Using Adaptive Equalization
Tile: 4 rows x 4 columns
Augmentations
Outputs per training example: 5
Flip: Horizontal
Saturation: Between -46% and +46%
Noise: Up to 5% of pixels
Cutout: 6 boxes with 33% size each
Bounding Box: Brightness: Between -61% and +61%
Bounding Box: Exposure: Between -55% and +55%
Bounding Box: Blur: Up to 10px