Summary

For Textile Industry, it is indispensable a proper assessment of yarns, since the quality of a product relies directly on the quality of the yarns.

Currently there are equipments able to perform yarn tests, namely the Uster Tester 5 and the Multitester da Zweigle [1, 2]. Despite their contribution for yarn characterization, these equipments have some disadvantages, such as, high cost, the equipments have large dimensions and a substantial weight, the resolution and accuracy limited in the mass parameters determination. With the actual state of computers evolution, it is believed that an Image Processing (IP) based technological solution, characterized by a high efficiency and feasibility, suppress all the identified drawbacks of traditional solutions [3, 4] and establish new standards of yarn parameterization [5].

Goals

This dissertation project seeks to develop a new technological solution to automatically characterize the yarn mass parameters, as well as, the detection of loop fibers, protruding fibers and the distinction between them, based on Image Processing (IP) techniques. It will also allow the possibility for online control during the production, due to the reduced measurement hardware required (an illumination source and digital camera with analogue optics coupled).

To achieve the objectives described, it is necessary to acquire yarn images, with different linear masses and characteristics, through a camera with analogue optics coupled. The samples images will be analyzed and processed using Labview software and Imaq Vision toolkit, both from National Instruments.

This solution will contribute to obtain a superior product quality and increase the efficiency of the textile industry.

Background

IP based applications have been used in the textile field since 1964 [6], although they have not been converted to viable quality control methods [7].

Due to its influence on the quality of textiles, yarn hairiness is considered to be one of the most significant parameters. In textile industry the equipments used to measure the hairiness are based on photoelectric methods, like the equipments Shirley's apparatus and the Uster Tester 3 [4]. Several algorithms are currently under development to characterize the yarn hairiness with IP. The method proposed by Marcin Kuzanski and Lidia Jackowska-Strumill, measure the real length of the protruding fibres from the yarn core and their number to measure the hairiness [5].

Anirban Guha, C. Amarenath, S. Pateria and R. Mittal [8] introduced a new method to measure the hairiness, based on the assumption that the hairs close and parallel to the yarn core would be a better indicator of hairiness, so a new concept it's created: Hair area index. This concept measure the area covered by hairs and divided by the area of the yarn core to obtain a dimensionless quantity. Still, it is necessary to develop algorithms to detect and characterize loop fibres length and to clearly distinguish between protruding fibres and loop fibres when they are interlaced

In textile industry, one of the most important parameter is the diameter, since it is used to predict fabric structural parameters such as width, cover factor, porosity and fabric comfort [9]. Some studies on diameter measurement were performed through image processing techniques. One of the techniques used is based on the extraction of the yarn core and the measure of the distance between edges is quantified. After the acquired data, it is made an average, calculating the yarn diameter [9, 10]. But, algorithms to measure the exact length of yarn irregularities (thin places, thick places and neps) [11,14] need to be developed. The characterization of yarn mass can be inferred from the yarn diameter, depending, among other parameters, on the yarn fibre density and porosity [12, 13].

Few algorithms to detect and distinguish between loop fibres and protruding fibres, are being developed.

Project Development

The project will be developed in the following phases:

     Study of the image processing techniques, so as the various parameters of a textile yarn.

    Research about the background and read related articles to the image processing techniques.

     Image acquisition of 100% cotton yarns, with different linear masses and characteristics for further analysis.

    Develop an algorithm through image processing techniques to measure the yarn diameter. Perform tests to validate the algorithm. Write an article.

    Develop an algorithm through image processing techniques to measure the hairiness of a yarn. Test and validate the algorithm with different yarns. Write an article.

    Build an algorithm through image processing techniques to detect loop fibres and protruding fibres and the distinction between them. Perform tests to validate the algorithm.

     Thesis write.

References

• [1] url:www.uster.com (access in October 2009)
• [2] url:www.zweigle.com (access in October 2009)
• [3] M. Kuzanski and L. Jackowska-Strumillo, "Yarn Hairiness Determination - The algorithms of computer measurement methods", MEMSTECH'2007, May 23-26, Lviv-Polyana, Ukraine, p.154-157.
• [4] M. Kuzanski, "Measurement Methods for Yarn Hairiness Analysis - the idea and construction of research standing", MEMSTECH'2006, May 24-27, Lviv-Polyana, Ukraine, p.87-90.
• [5] Masajtis, J., "Thread image processing in the estimation of repetition of yarn Structure" Fibers and Textiles in Eastern Europe, July/September 1997.
• [6] Shiloh, M., "A computer for the evaluation of the crimp diameter of textile fibers" Textile Research Journal, 1964, 34(5): p. 430-434.
• [7] S. Jaganathan, "Characterization Methods and Physical Properties of Novelty Yarn" , MSc Thesis, North Carolina State University, Raleigh, 2005.
• [8] Anirban Guha, C. Amarenath, S. Pateria and R, "Measurement of yarn hairiness by digital image processing" The journal of the Textile Institute,2009,99999:1.
• [9] http://www.indiantextilejournal.com/articles/FAdetails.asp?id=552 (access in January 2010).
• [10] A. Basu, I. Doraiswamy, and R.L. Gotipamul, "Measurement of Yarn Diameter and Twist by Image Analysis",Journal of the Textile Institute, Volume 94, Issue 1 & 2 2003, p37-48
• [11] V. Carvalho, "Automatic Yarn Characterization System", PhD Thesis, Minho University, Braga,2008.
• [12] B.C. Goswami, J.G. Martindale and F.L. Scardino, Textile Yarns: Technology, Structure and Applications, 2nd Edition, John Wiley & Sons, New York, 1977.
• [13] A. Cay, S. Vassiliadis, M. Rangoussi and I. Tarakçioglu, "On the Use of Image Processing Techniques for the Estimation of the Porosity of Textile Fabrics", Proceedings of World Academy of Science, Engineering and Technology - PWASET, Vol. 2, January 2005, p.73-76.
• [14] V. Carvalho, "Parametrização de Fio Têxtil Baseada na Ánalise de Massa", M.Sc Thesis, Minho University, Braga 2004 (in portuguese).