• DocumentCode
    2960775
  • Title

    A novel object recognition method based on improved edge tracing for binary images

  • Author

    Talu, M. Fatih ; Turkoglu, Ibrahim

  • Author_Institution
    Dept. of Electron. & Comput. Sci., Firat Univ., Elazig, Turkey
  • fYear
    2009
  • fDate
    14-16 Oct. 2009
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    In this article, we mention about a novel object recognition method based on an improved edge tracing method for binary images. This method named as ¿NORMBI¿ (novel object recognition method in binary images) contains two subsequent processes, determining places of independent objects (the starting point detection algorithm) and obtaining their edge points and geometrical shapes information (the edge tracing algorithm). The most important difference between NORMBI and traditional methods is to obtain geometrical shapes information and edge points of objects simultaneously by scanning image matrix once. It means to improve speed factor that has a vital important for object recognition. Besides, a tool has been designed to obtain the performance differences in between NORMBI and traditional methods. The mathematical results obtained are represented comparatively in a tabular format.
  • Keywords
    edge detection; object recognition; NORMBI; binary images; geometrical shape information; image matrix scanning; improved edge tracing method; object recognition method; starting point detection algorithm; Computer science; Detection algorithms; Image edge detection; Image recognition; Informatics; Motion detection; Object detection; Object recognition; Robustness; Shape; Binary image; Edge tracing; Object recognition;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Application of Information and Communication Technologies, 2009. AICT 2009. International Conference on
  • Conference_Location
    Baku
  • Print_ISBN
    978-1-4244-4739-8
  • Electronic_ISBN
    978-1-4244-4740-4
  • Type

    conf

  • DOI
    10.1109/ICAICT.2009.5372611
  • Filename
    5372611