SURFACE INSPECTION OF TEXTILE COMPOSITE-MATERIALS USING IMAGE-ANALYSIS TECHNIQUES

Citation
Y. Gowayed et al., SURFACE INSPECTION OF TEXTILE COMPOSITE-MATERIALS USING IMAGE-ANALYSIS TECHNIQUES, Journal of composites technology & research, 18(1), 1996, pp. 3-14
Citations number
11
Categorie Soggetti
Polymer Sciences","Materials Sciences, Composites
ISSN journal
08846804
Volume
18
Issue
1
Year of publication
1996
Pages
3 - 14
Database
ISI
SICI code
0884-6804(1996)18:1<3:SIOTCU>2.0.ZU;2-X
Abstract
In this paper, a novel visual quality control technique for production of textile composite materials is presented. This technique is based on the idea of automated image processing. The approach starts by acqu iring images of the surface of a composite product using a combination of a microscope and a frame grabbing board connected to a computer. I mage processing is then applied to the acquired images to extract impo rtant features. In the current study, the features of importance are o rientation of surface yarns. The image analysis approach is constructe d taking into consideration the intrinsic characteristics of the acqui red images. Important image features are enhanced using a Difference o f Gaussians and a set of directional edge detection kernels. The enhan ced image is then thresholded using a ''Fixed Percent Setting'' techni que and converted into binary format. Hough Transform and Bounding Box approaches are utilized for object (that is, fibers) recognition in t he binarized image. This approach is carried out on 20 ceramic composi te parts. The fabric preform was manually placed in all these parts. T he technique was successful in determining predominate directions orie ntation of surface yarns in most of the parts. Matrix material in some areas of these parts were over-grown in the infiltration process. Due to this phenomena, 6% of the acquired images have no data available a bout yam's orientation. Consequently, the image analysis approach is u nsuccessful in obtaining information about yam directions in these are as.