Dataset Versions

v4

ver04

Generated on Jan 22, 2024

Dataset Split

Train Set 76%
260Images
Valid Set 12%
42Images
Test Set 12%
42Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization

Augmentations

Outputs per training example: 2
Rotation: Between -7° and +7°
Brightness: Between -10% and +10%