Fabric Defect

Fabric defect detection is a crucial area of research aiming to automate quality control in textile manufacturing and improve textile recycling processes. Current research focuses on developing robust unsupervised anomaly detection methods, employing convolutional neural networks, autoencoders, and student-teacher networks, often enhanced by techniques like knowledge distillation and feature selection to improve accuracy and speed. These advancements are driven by the need for efficient, real-time inspection systems capable of handling diverse fabric types and defect variations, ultimately leading to increased productivity, reduced waste, and improved sustainability in the textile industry.