U-net Dataset (Semantic segmentation)

Semantic Segmentation

Dataset Versions

v10

U-net_7_classes_Ballanced

Generated on Mar 3, 2024

Dataset Split

Train Set 85%
1262Images
Valid Set 9%
126Images
Test Set 6%
93Images

Preprocessing

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

Augmentations

Outputs per training example: 2
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise