Dataset Versions

v4

2023-05-12 4:30pm

Generated on May 12, 2023

Dataset Split

Train Set 97%
3188Images
Valid Set 2%
60Images
Test Set 1%
26Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 1280x1280

Augmentations

Outputs per training example: 2
Flip: Horizontal, Vertical
Rotation: Between -45° and +45°
Shear: ±25° Horizontal, ±25° Vertical
Saturation: Between -32% and +32%
Brightness: Between -64% and +64%
Exposure: Between -36% and +36%
Blur: Up to 1.75px
Noise: Up to 6% of pixels
Mosaic: Applied
Bounding Box: Flip: Horizontal, Vertical

Similar Projects

See More