A Direct Measurement Method of Yarn Evenness Based on Machine Vision. Issue 4 (December 2015)

• Record Type: Journal Article
• Authors: Li, Junjuan; Zuo, Baoqi; Wang, Chen; Tu, Wenxiao
• Is Part Of: Journal of engineered fibers and fabrics. Volume 10:Issue 4(2015)
• Publication Date: 2015-12
• Subjects: Yarn evenness -- machine vision -- pictorial indices -- image processing; Nonwoven fabrics -- Periodicals; Fibers -- Periodicals; Fibers; Nonwoven fabrics; Periodicals; 677.6
• DOI: 10.1177/155892501501000403
• Languages: English
• ISSNs: 1558-9250

Abstract

In this paper, a new yarn evenness measurement method based on machine vision is introduced, which is a direct measurement process, as opposed to other methods. Two types of yarns (i.e., same yarn count but different quality grade and same quality grade but different yarn count) are measured to determine the coefficient of variation unevenness, which can be compared with the results of USTER ME100. The yarn images are continuously captured via an image acquisition system. To determine the main body of the yarn accurately, the yarn images are processed sequentially by a threshold segmentation and morphological opening operation. Next, the coefficient of variation (CV value) of the diameter is calculated to characterize the yarn evenness. Different image processing methods are used and compared to obtain a suitable method for use in the experiment. A more accurate, more efficient, and faster measurement system will meet requirements of the manufacturing of yarn; the suitable performance of the proposed method is illustrated using experimental results.