Computer-aided image processing method for yarn hairiness evaluation

Indexed in SCOPUS Yarn hairiness is one of the most important parameter to be measured in textile industry, since it affects aesthetical and technical features of yarns and, consequently, of fabrics. Commercial devices for evaluating yarn hairiness are commercially available; these are characterized by a number of drawbacks such as high cost, considerable dimensions and weight. In order to overcome these limitations a number of works have been developed aiming to measure the hairiness by using image processing based methods. Most of these works propose a hairiness coefficient defined as the ratio between the total length of fibers protruding from the yarn body and a reference length. On the basis of these works, in the present paper a new methodology for yarn hairiness estimation based on a computer-aided image processing approach is presented. In particular, a simple acquisition system is proposed with the aim of determining a characteristic curve of the yarn hairiness and of defining an “equivalent yarn hairiness”. The approach has been tested against the traditional Uster tester which is commonly accepted as one of the most important methods for measuring hairiness. Results show that the proposed method is suitable for the evaluation of yarn hairiness with an average error of about 7.5%.