International Research Journal of Education and Technology

The textile industry faces significant challenges in maintaining high-quality standards due to frequent occurrences of fabric defects. Traditional manual inspection methods are often inefficient and error-prone. This project presents an automated solution using artificial neural networks (ANN) to identify and classify fabric defects. By leveraging the ZJU Leaper Fabric Dataset, this study utilizes image preprocessing, model training, and performance evaluation to develop a robust ANN-based defect detection system. The outcomes demonstrate the ANN's efficacy, achieving notable performance metrics and offering substantial improvements over manual and existing automated methods. This research contributes to efficient quality control in textile manufacturing .The textile industry faces significant challenges in maintaining high-quality standards due to frequent occurrences of fabric defects. Traditional manual inspection methods are often inefficient and error-prone. This project presents an automated solution using artificial neural networks (ANN) to identify and classify fabric defects. By leveraging the ZJU-Leaper Fabric Dataset, this study utilizes image preprocessing, model training, and performance evaluation to develop a robust ANN-based defect detection system. The outcomes demonstrate the ANN's efficacy, achieving notable performance metrics and offering substantial improvements over manual and existing automated methods. This research contributes to efficient quality control in textile manufacturing.

Keywords – Textile industry,Fabric defect detection,Artificial neural networks (ANN),Automated quality control,ZJU-Leaper Fabric Dataset,Image preprocessing,Model training,Performance evaluation,Defect classification,Quality standards in textile manufacturing