Akademik Veri Yönetim Sistemi
Journal of the Textile Institute, 2025 (SCI-Expanded, Scopus)
Image analysis and computer vision are becoming important tools for studying different yarn characteristics. This review explores how these techniques are applied to measure properties such as yarn diameter, twist, porosity, hairiness, unevenness, and even internal structure. Compared to traditional methods, image analysis is faster, more precise, non-destructive and often more cost-effective as well as environmentally friendly. Several methods, such as Canny edge detection, Hough transforms, clustering approaches and machine learning models, have been used to study and measure these yarn properties. In this review, these algorithms and their roles in yarn analysis have been discussed. The discussion also brings together important studies in the field, points out the challenges researchers are still facing and suggests possible ways forward to improving yarn characterization with computer vision and image analysis.