• DocumentCode
    2313930
  • Title

    Edge detection of cord fabric defects image based on an improved morphological erosion detection methods

  • Author

    Wang, Dongyun ; Liu, Huiqin

  • Author_Institution
    Electr. Dept., Zhongyuan Univ. of Technol., Zhengzhou, China
  • Volume
    8
  • fYear
    2010
  • fDate
    10-12 Aug. 2010
  • Firstpage
    3943
  • Lastpage
    3947
  • Abstract
    An improved erosion detection method is proposed in this paper for edge detection of defect image Pulp plaque which is processed by improved morphological filtering. Furthermore, applying some traditional methods, such as difference marginal check method, Laplace edge check method, Sobel edge check method, morphological dilation detection method and morphology of erosion detection method, to detect the image edge. The experiment results show that improvement shape erosion examination algorithm proposed by this article has the noise elimination influence function, and furthermore, can retain the edge detail information of the goal in enormous degree, which is a effective edge detection method.
  • Keywords
    edge detection; fabrics; filtering theory; image denoising; image segmentation; matrix algebra; Laplace edge check method; Sobel edge check method; cord fabric defect image; defect image pulp plaque; difference marginal check method; image edge detection; image segmentation; improved morphological erosion detection methods; improved morphological filtering; morphological dilation detection method; noise elimination; shape erosion examination algorithm; Corrosion; Digital images; Image edge detection; Laplace equations; Morphology; Noise; Shape; cord fabric; defects image; edge detection; morphological erosion;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Natural Computation (ICNC), 2010 Sixth International Conference on
  • Conference_Location
    Yantai, Shandong
  • Print_ISBN
    978-1-4244-5958-2
  • Type

    conf

  • DOI
    10.1109/ICNC.2010.5584778
  • Filename
    5584778