Fully Automated Fabric Defect Detection Using Additive Wavelet Transform

This paper introduces a proposed fabric defect detection technique based on additive wavelet transform. In this paper, à trous wavelet is utilized to extract the approximate sub image at an appropriate level. The objective of the proposed technique is to enhance energy of defective region and attenuate energy of background in the selected level. An improved thresholding method based on statistical calculation is used. © 2020, Menoufia University, Faculty of Electronic Engineering. All rights reserved.