Extraction of Kannada text from different sets of Images

  Leena A,  Pooja A P

Extraction of text content has raised immense thought from the Computer vision group as of late. Extraction process is said to be tough in images that is taken by camera. Extraction of content is a vital assignment because of it wide mixed bag of uses, for example, route, human Computer association, substance primarily based data indexing and recovery, image comprehension, image pursuit and indexing, programmed annotation and organizing of images. Significantly extraction of Kannada content from pictures is tough because of geometry and between characters separating which is non-uniform in Kannada script. The primary point of the work is to develop a system for extraction of Kannada text and to expel the non message area from shaded complex foundation pictures. So content can be seen clearly, that is helpful for a few applications. Therefore concentrating on the Kannada text from images possible with complicated foundation here a basic and effective Curvelet Transform is utilized. It changes an image by reducing it into an arrangement of directional sub-groups with details of surface captured in distinctive orientations and scales. The algorithms are executed in matlab and applied to an image set with distinctive text size, text designs and thickness.

Keywords- Curvelettransform,canny,dilation,clustering.

Unique Identification Number - IJEDR1503046
Page Number(s) -
Pubished in - Volume 3 | Issue 3 | July 2015
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Leena A,  Pooja A P,   "Extraction of Kannada text from different sets of Images", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 3, pp., July 2015, Available at :http://www.ijedr.org/papers/IJEDR1503046.pdf