PCA Based English Handwritten Digit Recognition

In this paper, a digit Recognition system is designed using the Principal Component Analysis (PCA), a method of extraction of characteristics based on the digit forms, combined with kNearest Neighbor to recognize the numeral digits, this approach is tested on the MNIST handwritten isolated digit dat...

Full description

Saved in:
Bibliographic Details
Published in:International journal of advanced research in computer science 2017-05, Vol.8 (5), p.1426
Main Authors: Yogish Naik G R, Amani Ali Ahmed Ali
Format: Article
Language:English
Subjects:
Handwriting recognition
Principal components analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, a digit Recognition system is designed using the Principal Component Analysis (PCA), a method of extraction of characteristics based on the digit forms, combined with kNearest Neighbor to recognize the numeral digits, this approach is tested on the MNIST handwritten isolated digit database. This proposed method shows an excellent performance with higher accuracy, Achieved approximately 86.5%.
ISSN:0976-5697
DOI:10.26483/ijarcs.v8i5.3253