In situ characterization of the morphological wrinkling of woven fibrous materials by a mechanical test

Abstract

Wrinkling is one of the most common flaws of woven fabrics in domestic use and industrial applications. It is necessary to develop an objective evaluation method to quantify the smoothness appearance of fabrics effectively. Herein, a fabric multi-deformation tester (FMDT) was designed to evaluate the smoothness appearance of garment fabrics by one sequential mechanical test, overcoming the main difficulties of the existing visual measurement methods for wrinkling evaluation of fabrics with complex colors and patterns. The k-means clustering algorithm was used to objectively cluster the fabric samples based on the characteristic parameters, including the wrinkle-induced residual force ( F wr), hysteresis distance ( H fr), position deflection ( D fr) and stretching recovery slope ( S tr), from the testing curve and the thickness and weight of fabrics, and comparisons with subjective evaluation were also conducted. The results reveal that the k-means clustering is able to classify the smoothness appearance of fabrics using the selected characteristic parameters, showing a good consistency with the subjective clustering results. The feasibility of using the mechanical and deformation properties of textiles to characterize fabric smoothness appearance is proved, and the FMDT provides a potential method to analyze the wrinkling of fibrous materials in a convenient way.

Fusing Convolutional Neural Network Features With Hand-Crafted Features for Objective Fabric Smoothness Appearance Assessment

“…In addition, there is a series of methods that does not measure the wrinkles on the sample surface, namely indirect measurement. These methods evaluate the fabric wrinkle property by measuring the fabric weave parameters, mechanical properties [49], in-situ characterization [50], and shape retention properties [51] of the fabric sample and mapping to the smoothness appearance degree or wrinkle recovery angle.…”