The inspection of fabric defects by using wavelet transform

Authors
Citation
Mc. Hu et Is. Tsai, The inspection of fabric defects by using wavelet transform, J TEXTILE I, 91(3), 2000, pp. 420-433
Citations number
10
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Material Science & Engineering
Volume
91
Issue
3
Year of publication
2000
Part
1
Pages
420 - 433
Database
ISI
SICI code
Abstract
in this study, wavelet transform and an artificial neural network (ANN) are used to inspect four kinds of fabric defect, Multiresolution representatio n of an image by using wavelet transform is a new and effective approach to the analysis of image information content. The transform can be computed e fficiently by a pyramidal algorithm. The result is a set of sub-band images that consist of a lower-resolution version of the original image and a seq uence of detail sub-images containing higher-spectral information. Since th e transform generates localized spatial and frequency information simultane ously, the location and the kind of fabric defect can be inspected. In this study, we calculate the average and standard deviation for each sub-image as feature parameters for ANN. We explore three basic considerations on the classification rate of fabric-defect inspection consisting of wavelets wit h various maximum vanishing moments, different numbers of resolution levels , and different scaled-fabric images. The results show that the total class ification rate for a wavelet function with a maximum vanishing moment of 4 and resolution levels of 3 can reach 100%, and the different-scaled fabric image does not obviously affect the classification rate.