Dataset Versions
v3
2024-04-26 4:16pm
Generated on Apr 26, 2024
160 Total Images
View All Images
Dataset Split
Train Set 88%
141Images
Valid Set 8%
12Images
Test Set 4%
7Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Contrast Stretching
Augmentations
Outputs per training example: 3
Rotation: Between -3° and +3°
Grayscale: Apply to 15% of images
Hue: Between -21° and +21°
Blur: Up to 2.5px
Noise: Up to 6.04% of pixels
Cutout: 8 boxes with 12% size each
Mosaic: Applied