Improved Handwritten Digit Recognition method using Deep Learning Algorithm

Ruslan Jantayev, Yedilkhan N. Amirgaliyev

2019

Closed access Metadata (CSL-JSON)
DOI 10.1109/icecco48375.2019.9043235
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

One of the essential problems in Computer Vision is identification and classification of important objects. While exhaustive work done on image processing for computation and accuracy performance it is still limited by ambiguity. In current work we compared traditional machine learning method versus Deep Learning model, namely Convolutional Neural Network(CNN), on Handwritten Digit Recognition using MNIST dataset. We showed that CNN algorithm reaches higher recognition accuracy than Support Vector Machine(SVM).