Udine University

crackkkkk

Instance Segmentation

crackkkkk Image Dataset

v1

2023-04-13 1:43pm

Generated on Apr 13, 2023

Dataset Split

Train Set 87%
8449Images
Valid Set 8%
806Images
Test Set 4%
403Images

Preprocessing

Auto-Orient: Applied
Static Crop: 25-75% Horizontal Region, 25-75% Vertical Region
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Rotation: Between -6° and +6°
Blur: Up to 1.75px
Bounding Box: Rotation: Between -7° and +7°