A New Fabric Defect Detection Model based on Summed-up Distance Matching Function and Gabor Filter Bank

DOI

10.2991/jimec-18.2018.3How to use a DOI?

Keywords

periodic texture; Gabor; Binary Particle Swarm Optimization; defect detection

Abstract

Focusing on the problem of low efficiency and high mistake rate for manual detection of fabric with periodic textures, a new fabric defect detection model based on summed-up distance matching function and optimal Gabor filter bank is proposed. Firstly, the frequency and scale parameters of the filter bank are accurately calculated by the summed-up distance matching function; for the direction parameter, the objective function is constructed by the mean and variance of the energy of the normal fabric images convolved by the Gabor filter bank, then the optimal direction is determined by particle swarm algorithm. The optimal filter bank is used to convolute the fabric image to be measured, and then the Otsu algorithm is used to accurately segment the convolution image. The experimental results show the method has a short learning time and good robustness, and can detect fabric flaws quickly and accurately.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yihong Zhang
AU  - Xin Ruan
AU  - Shijie Pan
AU  - Lifeng Shi
AU  - Bao Zong
PY  - 2018/12
DA  - 2018/12
TI  - A New Fabric Defect Detection Model based on Summed-up Distance Matching Function and Gabor Filter Bank
BT  - Proceedings of the 2018 3rd Joint International Information Technology，Mechanical and Electronic Engineering Conference (JIMEC 2018)
PB  - Atlantis Press
SP  - 13
EP  - 16
SN  - 2589-4943
UR  - https://doi.org/10.2991/jimec-18.2018.3
DO  - 10.2991/jimec-18.2018.3
ID  - Zhang2018/12
ER  -