Defect Detection in Batavia and Sarga Woven Fabrics by Means of Convolutional Neural Networks

Quality assurance stands as a pivotal phase across all manufacturing pro- cesses, particularly within the textile industry. Presently, textile inspections heavily rely on human visual assessment due to deficiencies in commercial so- lutions. In response, this research advocates for the adoption of Deep Learn- ing models to automate fabric quality control within dynamic and contempo- rary production environments. Specifically, Convolutional Neural Networks are scrutinized using authentic images sourced from the Batavia and Sarga weave production lines. The experimental results identify DenseNet121 and Incep- tionV3 as the most effective models for Batavia and Sarga weaves, respectively. DenseNet121 demonstrates balanced performance across key metrics for Batavia weave, while InceptionV3 excels in Sarga weave, particularly in F1-score and AU-ROC. These findings underscore the potential of DL models to enhance the accuracy and efficiency of textile quality control.

Authors

• Beatriz Gil-Arroyo
• Nuria Velasco-Pérez
• Juan Marcos Sanz
• Abraham Casas
• Ángel Arroyo
• Daniel Urda

Keywords

• Textile defect detection
• Fabric quality control
• Deep Learning in manufacturing
• Convolutional neural networks
• Image analysis and clas-sification