Automated Math Symbol Classification Using SVM

Handwritten character/symbol recognition is an important area of research in the present digital world. The solving of problems such as recognizing handwritten characters/symbols written in different styles can make the human job easier. Mathematical expression recognition using machines has become...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of e-collaboration Vol. 18; no. 2; pp. 1 - 14
Main Authors Vaidehi K, Manivannan R
Format Journal Article
LanguageEnglish
Published Hershey IGI Global 01.01.2022
Subjects
Accuracy
Analysis
Automation
Character recognition
Classification
Collaboration
Employee motivation
Feature extraction
Fourier transforms
Handwriting
Handwriting recognition
Mathematical analysis
Neural networks
Support vector machines
India
Online AccessGet full text

Cover

More Information
Summary:Handwritten character/symbol recognition is an important area of research in the present digital world. The solving of problems such as recognizing handwritten characters/symbols written in different styles can make the human job easier. Mathematical expression recognition using machines has become a subject of serious research. The main motivation for this work is both recognizing of the handwritten mathematical symbol, digits and characters which will be used for mathematical expression recognition. The system first identifies the contour in handwritten document segmentation and features extracted are given into SVM classifier for classification. GLCM and Zernike Moments are the two different feature extraction techniques used in this work. SVM with RBF kernel is used for classification. Zernike Moment features overperforms than GLCM. Zernike Moment achieves 97.89% accuracy and GLCM achieves 87.61% accuracy.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1548-3673
1548-3681
DOI:10.4018/IJeC.304037