International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
Volume 48 - Issue 8 |
Published: June 2012 |
Authors: Ashutosh Aggarwal, Rajneesh Rani, Renu Dhir |
![]() |
Ashutosh Aggarwal, Rajneesh Rani, Renu Dhir . Recognition of Devanagari Handwritten Numerals using Gradient Features and SVM. International Journal of Computer Applications. 48, 8 (June 2012), 39-44. DOI=10.5120/7371-0151
@article{ 10.5120/7371-0151, author = { Ashutosh Aggarwal,Rajneesh Rani,Renu Dhir }, title = { Recognition of Devanagari Handwritten Numerals using Gradient Features and SVM }, journal = { International Journal of Computer Applications }, year = { 2012 }, volume = { 48 }, number = { 8 }, pages = { 39-44 }, doi = { 10.5120/7371-0151 }, publisher = { Foundation of Computer Science (FCS), NY, USA } }
%0 Journal Article %D 2012 %A Ashutosh Aggarwal %A Rajneesh Rani %A Renu Dhir %T Recognition of Devanagari Handwritten Numerals using Gradient Features and SVM%T %J International Journal of Computer Applications %V 48 %N 8 %P 39-44 %R 10.5120/7371-0151 %I Foundation of Computer Science (FCS), NY, USA
Recognition of Indian languages is a challenging problem. In Optical Character Recognition (OCR), acharacter or symbol to be recognized can be machine printed or handwritten characters/numerals. Several approaches in the past have been proposed that deal with problem of recognition of numerals/character depending on the type of feature extracted and way of extracting them. In this paper also a recognition system for isolated Handwritten Devanagari Numerals has been proposed. The proposed system is based on the division of sample image into sub-blocks and then in each sub-block Strength of Gradient is accumulated in 8 standard directions in which Gradient Direction is decomposed resulting in a feature vector with dimensionality of 200. Support Vector Machine (SVM) is used for classification. Accuracy of 99. 60% has been obtained by using standard dataset provided by ISI (Indian Statistical Institute) Kolkata.