mrFahrenhiet/CrackSegmentationDeepLearning

Multiscale Attention Based Efficient U-Net for Crack Segmentation, segments a RGB image into 2 classes crack and non-crack, this method obtained SOTA results on Crack500 dataset

26
/ 100
Experimental

This project helps civil engineers and maintenance crews automatically identify and monitor cracks in structures like pavements. It takes an RGB image of a surface and outputs a segmented image highlighting areas identified as cracks, distinguishing them from non-crack areas. This tool is designed for professionals involved in structural health monitoring and infrastructure inspection.

No commits in the last 6 months.

Use this if you need to precisely locate and map cracks within images for structural assessment and maintenance planning.

Not ideal if you need to detect other types of structural damage beyond cracks, or if your images are not clear enough to differentiate surface features.

structural-health-monitoring infrastructure-inspection pavement-maintenance condition-assessment civil-engineering
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

44

Forks

4

Language

Python

License

Last pushed

Oct 15, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mrFahrenhiet/CrackSegmentationDeepLearning"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.