Dataset Versions
v2
2024-08-07 8:40am
Generated on Aug 7, 2024
360 Total Images
View All Images
Dataset Split
Train Set 88%
315Images
Valid Set 8%
30Images
Test Set 4%
15Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 512x512
Auto-Adjust Contrast: Using Adaptive Equalization
Grayscale: Applied
Filter Null: Require at least 10% of images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Rotation: Between -20° and +20°
Shear: ±13° Horizontal, ±15° Vertical
Exposure: Between -13% and +13%
Blur: Up to 1.1px
Noise: Up to 1.41% of pixels
Cutout: 5 boxes with 7% size each