Advancements in Computer Vision and Deep Learning for Enhanced Textile Inspection and Manufacturing

There have been tremendous developments on the Technology front in recent times. New technologies, such as Artificial Intelligence (AI), the Internet of Things (IoT), Big Data Analytics, and Virtual Reality, have transformed many industries. The textile industry is also changing the way it operates. Technologies such as computer vision and deep learning have influenced how the textile industry functions. This study delves into integrating computer vision and deep learning in textile inspection, driving transformative shifts in defect detection, fabric pattern recognition, and real-time texture tracking. Notable contributions, such as Residual Networks for fabric pattern recognition and linear neural networks for fast web inspection, showcase the paradigm shift from manual to automated processes. Incorporating methodologies like CNN, GLCN, Gabor wavelet analysis, PCA, and innovative defect detection algorithms, the research explores diverse textures and regularity, emphasizing local and global assessments in woven fabric analysis. Results from topic modeling highlight key themes like textile defect detection and real-time fabric identification. The study concludes with managerial implications stressing the strategic adoption of automated systems and envisions future advancements in neural networks and real-time technologies, offering a comprehensive perspective on the current state, challenges, and prospects in automated textile inspection and manufacturing.