An Image Processing Research Consistent With Standard Photographs to Determine Pilling Grade of Woven Fabrics

Abstract

In this study, MATLAB 2018a software was used to evaluate pilling grade of woven fabrics objectively. Experimental works were carried out on the EMPA W3 standard photographs and accordingly two woven fabrics. Equations were built based on the measurements of pill characteristics and textural parameters of these photographs with the help of curve fitting method after image processing steps. Intervals were generated for each fabric by using slope of these equations and quantitative parameters obtained from the original fabric. Furthermore, fabrics that were provided pilling formation at different test turns were evaluated subjectively by expert operators.Objective results corresponding to each parameter were analyzed comparatively with these subjective results. The developed method was successful by using mean of matrix elements from textural parameters and total area from pill characteristics. ARTICLE HISTORY

The Comparison of the Edge Detection Methods in the Determination of Yarn Hairiness through Image Processing

“…Mean of matrix elements, standard deviation of matrix elements and entropy of grayscale imagesfrom textural parameters were used in a limited number of previous studies. Telli (2019) and Telli (2020) indicated that success has been achieved by using means of matrix elements in their developed method to determine the pilling grade of fabrics [27][28]. In this study, these three textural parameters showed strong correlation over 0.75 with all Zweigle indices in the Sobel and Prewitt methods.…”

“…After, these matrices in "1000x2500 uint8" format were transformed to double (1000x2500 double) format. A median filter was implemented on matrices in this format and image intensity values were adjusted for edge detection by using histogram-fitting [27,28]. Then, edge detection was applied in seven different methods.…”

“…Entropy is related to normalized histogram numbers. It is easily possible to reach from an open-access MATLAB database to the explanations and details related to the used edge detection methods and the performed functions for textural properties [27][28][30][31][32][33].…”

Investigation of Anti-Pilling Properties of Different Fabrics Applied with Polyvinylcaprolactam

“…A lot of parameters influence pilling in knitted and woven fabrics, like type of fibres, shape of fibres, fibre staple length, spinning technology used to produce the yarn, fabric construction, finishing technology, etc. All these parameters influence the pilling tendency of a garment [4]. Knitted fabrics tend to pill more readily than woven fabrics.…”

“…With the increase use of synthetic fibers and their blends in recent decades, the importance of pilling has increased even more [4]. In example polyester fabrics have excellent properties such as high strength, attractive hand, dimensional stability and easy-care properties.…”