Dataset Versions

v9

2023-01-11 5:19pm

Generated on Jan 11, 2023

Dataset Split

Train Set 86%
1020Images
Valid Set 7%
86Images
Test Set 7%
84Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 648x648
Modify Classes: 0 remapped, 0 dropped

Augmentations

Outputs per training example: 3
Flip: Horizontal
Rotation: Between -30° and +30°
Grayscale: Apply to 20% of images
Hue: Between -148° and +148°
Saturation: Between -90% and +90%
Brightness: Between -27% and +27%
Blur: Up to 3.25px
Noise: Up to 3% of pixels
Mosaic: Applied
Bounding Box: Brightness: Between -24% and +24%

Similar Projects

See More
572 images 2 models