In this paper a novel method was presented for evaluating the fabric pills using digital image processing techniques. This work provides a novel technique for detecting pills and also measuring their heights, surfaces and volumes. Surely, measuring the intensity of defects by human vision is an inaccurate method for quality control; as a result, this problem became a motivation for employing digital image processing techniques for detection of defects of fabric surface. In the former works, the systems were just limited to measuring of the surface of defects, but in the presented method the height and the volume of defects were also measured, which leads to a more accurate quality control. An algorithm was developed to first, find pills and then measure their average intensity by using three criteria of height, surface and volume. The results showed a meaningful relation between the number of rotations and the quality of pilled fabrics. Keywords—3D analysis, computer vision, fabric, pile, surface evaluation
[1]
J. Amirbayat,et al.
The Objective Assessment of Fabric Pilling Part II: Experimental Work
,
1994
.
[2]
Bugao Xu,et al.
Instrumental Evaluation of Fabric Pilling
,
1997
.
[3]
Koichiro Toriumi,et al.
Evaluation of Pilling by Computer Image Analysis
,
1988
.
[4]
J. Amirbayat,et al.
The Objective Assessment of Fabric Pilling. Part I: Methodology
,
1993
.
[5]
Christopher G. Harris,et al.
A Combined Corner and Edge Detector
,
1988,
Alvey Vision Conference.
[6]
Renyi,et al.
Towards objective visual assessment of fabric features
,
1999
.
[7]
Hazem M. Abbas,et al.
Automated vision system for localizing structural defects in textile fabrics
,
2005,
Pattern Recognit. Lett..