Automated concrete crack detection enhanced by deep learning and generative adversarial networks

Yue Chang · 2026

Read Automated concrete crack detection enhanced by deep learning and generative adversarial networks by Yue Chang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0354018.txt.

Description

Yue Chang proposes an automated way to find and measure cracks in concrete, published in PLOS ONE in 2026. Human inspectors judge small cracks inconsistently and sometimes at personal risk, while computer methods starve for labelled training images. The author's answer pairs an image-generating network, which fabricates extra realistic crack photos, with a segmentation network sharpened by attention modules to outline real cracks in new pictures.

Trained this way, the model outlined cracks with accuracy and F1 scores in the low eighties. A fractal measurement step then quantified each crack's geometry. The pipeline points toward cheaper, safer routine checks of bridges, dams, and towers, where catching a crack early prevents costly failures.

Catalog record

CREATORYue Chang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE47778
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0354018

Browse the full library · About the Tower of Babel