• DocumentCode
    1583844
  • Title

    Challenges of broken characters in character segmentation method for Gujarati printed documents

  • Author

    Shah, Riddhi J. ; Ratanpara, Tushar V.

  • Author_Institution
    Dept. of Comput. Eng., Dharmsinh Desai Univ., Nadiad, India
  • fYear
    2015
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    Character segmentation is an important step of optical character recognition (OCR) system. The segmentation of broken characters is one of the key factors which affect the performance of recognition system. So it is necessary to make OCR systems more effective and accurate for broken character segmentation. This paper explains the challenges in broken character segmentation method for Guajarati printed documents. Features and limitations of various methods in broken character segmentation are discussed. Different types of broken characters in Guajarati language are explained. Several challenges are explained for Gujarati broken character segmentation which is useful to implement broken character segmentation method.
  • Keywords
    image segmentation; optical character recognition; Guajarati printed documents; Gujarati printed documents; broken character segmentation; broken characters segmentation; optical character recognition; recognition system; Accuracy; Character recognition; Conferences; Image recognition; Image restoration; Image segmentation; Optical character recognition software; Broken characters; Optical character recognition; Recognition rate; Segmentation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Innovations in Information, Embedded and Communication Systems (ICIIECS), 2015 International Conference on
  • Conference_Location
    Coimbatore
  • Print_ISBN
    978-1-4799-6817-6
  • Type

    conf

  • DOI
    10.1109/ICIIECS.2015.7193263
  • Filename
    7193263