Performance Evaluation for Motif-Based Patterned Texture Defect Detection

doi:10.1109/TASE.2008.2005418

Performance Evaluation for Motif-Based Patterned Texture Defect Detection

2010-01-01

52

PRE

AI

DOI：10.1109/TASE.2008.2005418

Original

Original request for help

Save

Abstract

En 中文

This paper carries an extensive evaluation on the performance of a generalized motif-based method for detecting defects in 16 out of 17 wallpaper groups in 2-D patterned texture. The motif-based method evolves from the concept that every wallpaper group is defined by a lattice, which contains a further constituent-motif. It utilizes the symmetry properties of motifs to calculate the energy of moving subtraction and its variance among motifs. Decision boundaries are determined by learning the distribution of those values among the defect-free and defective patterns in the energy-variance space. In this paper, shape transform for irregular motif has been demonstrated according to the three basic motif shapes: rectangle, triangle, and parallelogram. An error analysis for the misclassifications has also been delivered. In the database of fabrics and other patterned textures, a total of 381 defect-free lattices are used for formulation of boundaries while further 340 defect-free and 233 defective lattices are for testing. The motif-based method has a consistent result and reaches a detection success rate of 93.86%. Note to Practitioners-This paper is motivated by the need to develop a generalized approach that can detect defects on most of the 2-D patterned textures defined so far. It proposes a novel motif-based defect detection method for 16 out of 17 wallpaper groups. A new concept called energy of moving subtraction is defined using norm metric measurement between a collection of circular shift matrices of motif and itself. Together with its variance, an energy-variance space is defined where decision boundaries are drawn for classifying defective and defect-free motifs. The method has been evaluated by two categories of patterned textures. The first category is produced from patterned fabric samples from p2, pmm, p4m, pm, and cm groups. The second category is produced from various patterned texture samples from p4, pg, pmg, cmm, p4g, pgg, p31m, p6, p6m, and p3m1 groups. For the former, a total of 280 defect-free lattices samples are used for deriving the decision boundaries, and further 340 defect-free and 206 defective lattices are used for evaluation. The detection success rate is found to be 93.92%. For the latter, they are the images from painting, tile, ornament, painted porcelain, vessel, earthenware, mat, tapestry, cloth, and wall tiling. A total of 101 defect-free lattices are acquired and 27 defective lattices are used for defect detection. The detection success rate for the second category is 92.59%. An overall detection success rate of 93.86% is achieved for the motif-based method. No other ( generalized) approach was able to handle such a large number of wallpaper groups of 2-D patterned textures, and hence this result outperforms all other previously published approaches. This result contributes to the quality assurance of production of textile, wallpaper, ceramics, ornament, and tile.

Keywords:

Defect detection

motif

patterned fabric

texture analysis

wallpaper group

Performance Evaluation for Motif-Based Patterned Texture Defect Detection

Abstract

AI Summary

Journal

Organization

Cited Papers