Computer vision techniques for detecting yarn defects

By Pereira, F.; Carvalho, V.; Soares, F.; Vasconcelos, R.; Machado, J.

Applications of Computer Vision in Fashion and Textiles

2018

Abstract

In textile industry, the quality of the final product is directly related to the quality of the yarn; therefore, it is essential to make an accurate assessment of the characteristics of the yarns, according to certain preestablished parameters. The main purpose of this chapter is to give an understanding of different problems in textile industry in what concerns quality and to take account of the main studies carried out to develop new technologies for assessing the quality of the yarn. Hence, different devices were analyzed to evaluate the quality of the yarn, presenting their advantages and limitations (e.g., the Tester 5, from USTER and the OASYS, from Zweigle (acquired by USTER)). Moreover, new solutions are proposed, based on Image-Processing Techniques that automatically characterize the mass parameters of the yarn with high reliability and efficiency and on artificial neural network models. The main gap observed is that most of the studies and developed systems focus only on analyzing a single textile yarn defect. There are some exceptions that we can highlight like the Yarn System Quality (YSQ) prototype and a recent computerized system to measure different yarn parameters. The YSQ prototype, for quality control of the yarn characteristics under laboratory conditions, can perform yarn periodical errors analysis, quantify the statistical yarn parameters, and establish a reliable method of yarn characterization by using direct measurements and new signal-processing approaches. The computerized system can give the mean diameter of the yarn under test in real-time units (millimeters), which means that hairiness is more meticulously defined. Furthermore, the system is immune to variations of ambient temperature, humidity, and illumination level.So, despite the significant progress of technology in the last decade, there is still room for improvement in yarn quality assessment systems to reduce the equipment cost and achieve a high product quality and production efficie

RepositoriUM:

Google Scholar: