Segmentation Framework for Multi-Oriented Text Detection and Recognition

  Shashi Kant,  Sini Shibu

Here in this paper a new and efficient technique for the text detection from natural scenes is implemented. The proposed methodology is based on the concept of Otsu’s segmentation method which segments the higher intensity texts from the natural scenes. Although there are various text detection techniques implemented, but the proposed methodology implemented here for text detection provides high precision and recall rate as compared to the existing technique. The methodology also provides less computational time to detect texts from scenes.

Keywords-

Unique Identification Number - IJEDR1504046
Page Number(s) - 308-315
Pubished in - Volume 3 | Issue 4 | November 2015
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Shashi Kant,  Sini Shibu,   "Segmentation Framework for Multi-Oriented Text Detection and Recognition", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 4, pp.308-315, November 2015, Available at :http://www.ijedr.org/papers/IJEDR1504046.pdf