Welcome to the InfoSci Platform
Reference Hub14
Boosting of Deep Convolutional Architectures for Arabic Handwriting Recognition

Boosting of Deep Convolutional Architectures for Arabic Handwriting Recognition

Mohamed Elleuch, Monji Kherallah
Copyright: © 2019 |Volume: 10 |Issue: 4 |Pages: 20
ISSN: 1947-8534|EISSN: 1947-8542|EISBN13: 9781522565352|DOI: 10.4018/IJMDEM.2019100102
Cite Article Cite Article

MLA

Elleuch, Mohamed, and Monji Kherallah. "Boosting of Deep Convolutional Architectures for Arabic Handwriting Recognition." IJMDEM vol.10, no.4 2019: pp.26-45. http://doi.org/10.4018/IJMDEM.2019100102

APA

Elleuch, M. & Kherallah, M. (2019). Boosting of Deep Convolutional Architectures for Arabic Handwriting Recognition. International Journal of Multimedia Data Engineering and Management (IJMDEM), 10(4), 26-45. http://doi.org/10.4018/IJMDEM.2019100102

Chicago

Elleuch, Mohamed, and Monji Kherallah. "Boosting of Deep Convolutional Architectures for Arabic Handwriting Recognition," International Journal of Multimedia Data Engineering and Management (IJMDEM) 10, no.4: 26-45. http://doi.org/10.4018/IJMDEM.2019100102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

In recent years, deep learning (DL) based systems have become very popular for constructing hierarchical representations from unlabeled data. Moreover, DL approaches have been shown to exceed foregoing state of the art machine learning models in various areas, by pattern recognition being one of the more important cases. This paper applies Convolutional Deep Belief Networks (CDBN) to textual image data containing Arabic handwritten script (AHS) and evaluated it on two different databases characterized by the low/high-dimension property. In addition to the benefits provided by deep networks, the system is protected against over-fitting. Experimentally, the authors demonstrated that the extracted features are effective for handwritten character recognition and show very good performance comparable to the state of the art on handwritten text recognition. Yet using Dropout, the proposed CDBN architectures achieved a promising accuracy rates of 91.55% and 98.86% when applied to IFN/ENIT and HACDB databases, respectively.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.