Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Multiple classifiers fusion and CNN feature extraction for handwritten digits recognition

Zhao, Huihuang and Liu, Han ORCID: https://orcid.org/0000-0002-7731-8258 2020. Multiple classifiers fusion and CNN feature extraction for handwritten digits recognition. Granular Computing 5 , pp. 411-418. 10.1007/s41066-019-00158-6

[thumbnail of 10.1007_s41066-019-00158-6.pdf]
Preview
PDF - Published Version
Available under License Creative Commons Attribution.

Download (854kB) | Preview

Abstract

Handwritten digits recognition has been treated as a multi-class classification problem in the machine learning context, where each of the ten digits (0-9) is viewed as a class and the machine learning task is essentially to train a classifier that can effectively discriminate the ten classes. In practice, it is very usual that the performance of a single classifier trained by using a standard learning algorithm is varied on different data sets, which indicates that the same learning algorithm may train strong classifiers on some data sets but weak classifiers may be trained on other data sets. It is also possible that the same classifier shows different performance on different test sets, especially when considering the case that image instances can be highly diverse due to the different handwriting styles of different people on the same digits. In order to address the above issue, development of ensemble learning approaches have been very necessary to improve the overall performance and make the performance more stable on different data sets. In this paper, we propose a framework that involves CNN based feature extraction from the MINST data set and algebraic fusion of multiple classifiers trained on different feature sets, which are prepared through feature selection applied to the original feature set extracted using CNN. The experimental results show that the classifiers fusion can achieve the classification accuracy of ≥ 98%.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Computer Science & Informatics
Publisher: Springer
ISSN: 2364-4974
Funders: National Natural Science Foundation of China, Cardiff University
Date of First Compliant Deposit: 26 February 2019
Date of Acceptance: 15 February 2019
Last Modified: 06 May 2023 05:44
URI: https://orca.cardiff.ac.uk/id/eprint/119813

Citation Data

Cited 54 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics