UGC Approved Journal no 63975(19)
New UGC Peer-Reviewed Rules

ISSN: 2349-5162 | ESTD Year : 2014
Volume 12 | Issue 9 | September 2025

JETIREXPLORE- Search Thousands of research papers



WhatsApp Contact
Click Here

Published in:

Volume 6 Issue 1
January-2019
eISSN: 2349-5162

UGC and ISSN approved 7.95 impact factor UGC Approved Journal no 63975

7.95 impact factor calculated by Google scholar

Unique Identifier

Published Paper ID:
JETIR1901E07


Registration ID:
306442

Page Number

42-51

Share This Article


Jetir RMS

Title

Recognition of Devanagari CAPTCHA Script that comprises characters with Middle-bar, No-bar and Digits using SVM and PNN

Abstract

Recognition of Devanagari CAPTCHA script characters is yet a challenging problem. Many lingual character recognition systems have been developed since last few years. There are several techniques that are proposed to deal with problem of character recognition. This paper presents an efficient Devanagari character recognition scheme that implements effective feature extraction methods and use classifiers like SVM (Support Vector Machine) and PNN (Probabilistic Neural Network) for recognizing printed Devanagari characters. To prepare a database, 5 samples of each Devanagari character from 5 different Kruti Dev fonts have been considered. There are 34 consonants and 13 vowels in Devanagari script. These characters are classified based on their structural properties. The presence or absence of vertical bar plays vital role in the structure of Devanagari characters. The vertical bar is used in the middle of some characters. It is also used as a terminator of many characters. There are 2 consonants that have middle-bar, 9 consonants and 3 vowels that have no-bar. There are 9 consonants out of 34 and 3 vowels out of 13, which have no end-bar. Such specific characters and 10 numeric characters are chosen for experiments. The experiments are performed on a dataset having 14 alphabetic characters (which have middle-bar & no-bar) and 10 numeric characters together, using SVM and PNN classifiers respectively. So, in all 24 characters are used to prepare a dataset of 6000 (24 x 250) character images. The proposed scheme has given average character recognition rates of 96.47% using SVM and 97.54%using PNN which are comparatively higher than other techniques.

Key Words

Keywords: SVM, PNN, Kruti Dev, Middle-bar characters, No-bar characters, End-bar characters, Dataset.

Cite This Article

"Recognition of Devanagari CAPTCHA Script that comprises characters with Middle-bar, No-bar and Digits using SVM and PNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 1, page no.42-51, January-2019, Available :http://www.jetir.org/papers/JETIR1901E07.pdf

ISSN


2349-5162 | Impact Factor 7.95 Calculate by Google Scholar

An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 7.95 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator

Cite This Article

"Recognition of Devanagari CAPTCHA Script that comprises characters with Middle-bar, No-bar and Digits using SVM and PNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 1, page no. pp42-51, January-2019, Available at : http://www.jetir.org/papers/JETIR1901E07.pdf

Publication Details

Published Paper ID: JETIR1901E07
Registration ID: 306442
Published In: Volume 6 | Issue 1 | Year January-2019
DOI (Digital Object Identifier): http://doi.one/10.1729/Journal.25968
Page No: 42-51
Country: Khamgaon, Maharashtra, India .
Area: Science & Technology
ISSN Number: 2349-5162
Publisher: IJ Publication


Preview This Article


Downlaod

Click here for Article Preview

Download PDF

Downloads

0002988

Print This Page

Current Call For Paper

Jetir RMS