Automatic quality control of textile webs by image processing

Various techniques based on image processing are presented for the automatic quality control of textiles. General defects (shrinking, abrasion, etc.) are detected by using operations in the frequency domain. Local defects (broken threads, mispicks, double yarns, etc.) are detected using a method based on a multiscale and multiorientation Gabor scheme that imitates the visual coding in early human vision. Also pilling resistance is automatically evaluated in wear-and-tear fabrics by a new algorithm which combines operations in both the spatial and frequency domain.