Fabric defect detection using fine-tuned Yolo-12

Waqar Ahmad, Rehan Ashraf, Toqeer Mahmood, Tanveer Hussain, Muhammad Haris Abid · 2026

Read Fabric defect detection using fine-tuned Yolo-12 by Waqar Ahmad, Rehan Ashraf, Toqeer Mahmood, Tanveer Hussain, Muhammad Haris Abid (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-0353550.txt.

Description

Ahmad and colleagues, in a 2026 study, fine-tune a fast vision model to find defects in fabric as it is made. Flawed cloth is wasted money, and human inspection is slow and tiring. Mills also make plain and printed textiles side by side, so one detector must handle both. The team trained on cloth from a Pakistani mill and checked the model on three public data sets.

Their customized model, built on the twelfth version of the YOLO detector family, learned features aimed at varied defect shapes and used adaptive tricks to enrich training images. On the home mill's cloth it scored well across seven defect types, balancing catches and false alarms. On the public sets it again beat comparison methods, staying light enough for cheap computing. The authors present it as a universal inspector for mixed textile lines, ready for the factory floor rather than only the lab.

Catalog record

CREATORWaqar Ahmad, Rehan Ashraf, Toqeer Mahmood, Tanveer Hussain, Muhammad Haris Abid
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE75341
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0353550

Browse the full library · About the Tower of Babel