Fabric defect detection using undecimated wavelet transform

In this study, a new fabric defect detection algorithm base on undecimated wavelet transform is proposed. The selection scheme of wavelet decomposition scales is investigated to set the decomposition scales adaptively to the fabric texture. The objective of the scheme is to enhance the energy of defective region and attenuate the energy of non-defective region. And the performance of detection results with different number of decomposition scales is also discussed. A simple and computationally effective data fusion scheme combined with amplitudes division of wavelet coefficients is used to fuse data from multiple scales together. And several features based on defective energy estimation are extracted from fused image. By examining the extracted features the proposed algorithm can provide not only the location of defects but also some detailed information about them which can be used for defect recognition and classification. Experimental results of real fabric defects are provided to validate the effectiveness and robustness of the defect detection algorithm. © 2011 Asian Network for Scientific Information.