A new method for classification of woven structure for yarn-dyed fabric
Research output: Journal Publications and Reviews (RGC: 21, 22, 62) › 21_Publication in refereed journal › peer-review
Author(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 78-95 |
Journal / Publication | Textile Research Journal |
Volume | 84 |
Issue number | 1 |
Online published | 26 Jun 2013 |
Publication status | Published - Jan 2014 |
Externally published | Yes |
Link(s)
Abstract
The fabric weave pattern recognition process is a structure identification process that detects the yarn location as well as the yarn crossing structure in a woven fabric. A new local orientation feature is proposed for fabric structure detection by using high-resolution images. The detection process consists of two main steps. Firstly, the yarn location is detected through a series of image enhancement techniques and an edge-based projection method. Secondly, the yarn float is recognized with a local orientation detection approach based on Radon transform. Three kinds of yarn-dyed cotton fabrics are investigated in this study, including the single yarn, the double yarn, and the twisted yarn fabric. Experimental results and discussions demonstrate that the research method is effective in detecting fabric structure and yarn float even with long hairiness. © 2014, SAGE Publications. All rights reserved.
Research Area(s)
- crossing point, Fabric image processing, fiber orientation, Radon transform, woven pattern recognition, yarn float
Citation Format(s)
A new method for classification of woven structure for yarn-dyed fabric. / Zheng, Dejun; Han, Yu; Hu, Jin lian.
In: Textile Research Journal, Vol. 84, No. 1, 01.2014, p. 78-95.Research output: Journal Publications and Reviews (RGC: 21, 22, 62) › 21_Publication in refereed journal › peer-review