Method of assessment of structural properties of knitted fabrics based on image analysis

Abstract

A novel digital image processing technique was devised to assess fabric structure attributes and identify weave patterns using a Wiener filter. Six distinct groups of images were acquired and subjected to analysis, encompassing three fabric structures with two constructions each. The developed method segregated the fabric image into two components, one containing warp yarns and the other weft yarns. By delineating yarn boundaries, the fabric’s surface characteristics were assessed, aiding in the identification of interlacing regions to discern the fabric structure. The findings demonstrated proficiency in evaluating surface fabric attributes and discerning fabric structures, particularly for fabrics with identical warp and weft yarn colors. Furthermore, the approach provided a more precise assessment of yarn spacing and fabric cover factor compared to conventional analytical techniques employed for such evaluations.