• DocumentCode
    2860530
  • Title

    Development of a real-time machine vision system for detecting defeats of cord fabrics

  • Author

    Zhang, Wuyi ; ZHAO, Qiangsong ; Liao, Liang

  • Author_Institution
    Sch. of Electron. Inf., Zhongyuan Univ. of Technol., Zhengzhou, China
  • Volume
    12
  • fYear
    2010
  • fDate
    22-24 Oct. 2010
  • Abstract
    Automatic detection techniques based on machine vision can be used in fabric industry for quality control, which constantly pursues intelligent methods to replace human inspections of product. This work introduces the principle components of a real-time machine vision system for defeat detection of cord fabrics, which is usually a challenging task in practice. The work aims at solving some difficulties usually incurring in such kind of tasks. The design and implementation of the algorithm, software and hardware are introduced. Based on the Gabor wavelet techniques, the system can automatically detect regular texture defects. Our experiments show the proposed algorithm is favorably suited for detecting several types of cord fabric defects. The system testing has been carried in both on-line and off-line situations. The corresponding results show the system has good performance with high detection accuracy, quick response and strong robustness.
  • Keywords
    Gabor filters; computer vision; fabrics; quality control; real-time systems; textile industry; wavelet transforms; Gabor wavelet technique; automatic detection technique; cord fabrics defect detection; fabric industry; human inspection; offline situation; online situation; quality control; real time machine vision system; texture defect; Fabrics; Feature extraction; Gabor filters; Image segmentation; Inspection; Machine vision; Cord fabric; Defect detection; Gabor wavelet; Machine vision;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Application and System Modeling (ICCASM), 2010 International Conference on
  • Conference_Location
    Taiyuan
  • Print_ISBN
    978-1-4244-7235-2
  • Electronic_ISBN
    978-1-4244-7237-6
  • Type

    conf

  • DOI
    10.1109/ICCASM.2010.5622393
  • Filename
    5622393