• DocumentCode
    3126437
  • Title

    Ant clustering based text detection in natural scene images

  • Author

    Tomer, Pooja ; Goyal, Ankur

  • Author_Institution
    Inf. Technol., Samrat Ashok Technol. Inst., Vidisha, India
  • fYear
    2013
  • fDate
    4-6 July 2013
  • Firstpage
    1
  • Lastpage
    7
  • Abstract
    Text pictures in natural scene are necessary clues for several image-based applications like scene understanding, content-based image retrieval, helpful navigation, and automatic geocoding. However, locating text from a fancy atmosphere with multiple colors may be a difficult task. Although there are various techniques implemented for the detection of the text in natural scenes. The work implemented in paper(Text string detection from natural scenes by structural based partition and grouping) is based on Gradient based Partition and Color based Partition but the techniques implemented here provides less efficiency and the chances of error rate is more hence have less accuracy. But the technique implemented in this paper for the detection of texts in natural scenes is based on Ant based clustering. By using Ant clustering the performance of the technique for the detection of text has been increased.
  • Keywords
    ant colony optimisation; document image processing; error statistics; image colour analysis; natural scenes; object detection; pattern clustering; text analysis; ant clustering based text detection; automatic geocoding; color based partition; content-based image retrieval; error rate; gradient based partition; grouping; image-based applications; natural scene images; navigation; scene understanding; structural based partition; text location; text pictures; text string detection; Character recognition; Clustering algorithms; Image color analysis; Image edge detection; Optical character recognition software; Partitioning algorithms; Sorting; Automatic geocoding; Connected Component; Gradient based partition; ant clustering;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing, Communications and Networking Technologies (ICCCNT),2013 Fourth International Conference on
  • Conference_Location
    Tiruchengode
  • Print_ISBN
    978-1-4799-3925-1
  • Type

    conf

  • DOI
    10.1109/ICCCNT.2013.6726747
  • Filename
    6726747