Dataset Versions
v1
2023-03-24 3:30pm
Generated on Mar 24, 2023
1132 Total Images
View All Images
Dataset Split
Train Set 98%
1104Images
Valid Set 2%
20Images
Test Set 1%
8Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization
Tile: 2 rows x 2 columns
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -15° and +15°
Brightness: Between -34% and +34%
Blur: Up to 2px
Bounding Box: Flip: Horizontal, Vertical
Bounding Box: Rotation: Between -15° and +15°
Bounding Box: Brightness: Between -41% and +41%