Comparative Evaluation of Generative Augmentation Techniques for Imbalanced Concrete Damage Segmentation: Classical Methods, GANs, and DiffusionModels

Khatry, Kalyan

doi:10.5281/zenodo.19376795

Published April 2, 2026 | Version v1

Publication Open

Comparative Evaluation of Generative Augmentation Techniques for Imbalanced Concrete Damage Segmentation: Classical Methods, GANs, and DiffusionModels

Khatry, Kalyan¹

1. University of New Haven

Contributors

Researcher:

Khatry, Kalyan¹

1. University of New Haven

Key Words:
Concrete damage segmentation
Class imbalance
Generative augmentation
Diffusion models
Structural health monitoring
Autonomous visual inspection

Files

Comparative_Evaluation__Classical_MethodsGANsand_Diffusion_Models.pdf

Files (39.1 MB)

Name	Size	Download all
Comparative_Evaluation__Classical_Methods__GANs__and_Diffusion_Models.pdf md5:7605a7dfcc22f9ce4241de6a5bf26cb7	39.1 MB	Preview Download

Additional details

Programming language: Python

106

Views

Downloads

Show more details

	All versions	This version
Views	106	106
Downloads	79	79
Data volume	4.0 GB	4.0 GB

More info on how stats are collected....

DOI

Resource type

Publication

Publisher

Zenodo

License: Creative Commons Attribution 4.0 International

The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited. Read more

Technical metadata

Created: April 1, 2026
Modified: April 2, 2026

Comparative Evaluation of Generative Augmentation Techniques for Imbalanced Concrete Damage Segmentation: Classical Methods, GANs, and DiffusionModels

Authors/Creators

Contributors

Researcher:

Description

Files

Comparative_Evaluation__Classical_Methods__GANs__and_Diffusion_Models.pdf

Files (39.1 MB)

Additional details

Software

Comparative_Evaluation__Classical_MethodsGANsand_Diffusion_Models.pdf