Defect Detection of Patterned Fabric by Spectral Estimation Technique and Rough Set Classifier

Abstract

A novel method for patterned fabric defect detection and classification using spectral estimation technique and rough set theory is presented in this paper. Estimating Signal Parameter via Rotational Invariance Technique (ESPRIT) is firstly used to extract the pattern from the image of the patterned fabric. Then, the shape and location of the flawed areas are detected by comparing the pattern image and the source image. A rough set classifier is trained and tested to detect the types of defects in the patterned fabric image. Experimental results show that this method can successfully analyze and recognize oil warp and weft defects in patterned fabrics with nearly 96% success rate.