Fabric Defect Detection Using YOLO V8

Textile quality assurance during processing is accomplished through fabric defect detection which is among the most important processes. The conventional manual inspection techniques are time consuming and subjective, resulting to inefficiency in quality flow. In recent years, the capability of computer vision and deep learning techniques to automate defect detection jobs has been proven. This paper puts forward a new technique in fabric defect detection using YOLO V8 which is a dominant object detection network. YOLO V8 presents an upgrade of performance and energy efficiency compared to the previous ones, therefore, the improved YOLO V8 version becomes more appropriate for defect detection in textiles. The experimental data confirms the validity of the proposed method in successfully identifying different fabric kinds of defects. Thus, it is a scalable and efficient option for checking quality in textile manufacturing.