Synthesizing versus Augmentation for Arabic Word Recognition with Convolutional Neural Networks

Reem Alaasam, Berat Kurar Barakat, Jihad El-Sana

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

In this paper, we present a sub-word recognition method for historical Arabic manuscripts, using convolutional neural networks. We investigate the benefit of extending training set with synthetically created samples in comparison to augmentation. We show that annotating around ten pages of a manuscript and extending it, is sufficient for successful sub-word recognition in the whole manuscript. In addition, we show the contribution of using different combinations of training sets and compare their sub-word recognition performance in the whole manuscript.

Original languageEnglish
Title of host publication2nd IEEE International Workshop on Arabic and Derived Script Analysis and Recognition, ASAR 2018
PublisherInstitute of Electrical and Electronics Engineers
Pages114-118
Number of pages5
ISBN (Electronic)9781538614594
DOIs
StatePublished - 2 Oct 2018
Event2nd IEEE International Workshop on Arabic and Derived Script Analysis and Recognition, ASAR 2018 - London, United Kingdom
Duration: 12 Mar 201814 Mar 2018

Publication series

Name2nd IEEE International Workshop on Arabic and Derived Script Analysis and Recognition, ASAR 2018

Conference

Conference2nd IEEE International Workshop on Arabic and Derived Script Analysis and Recognition, ASAR 2018
Country/TerritoryUnited Kingdom
CityLondon
Period12/03/1814/03/18

Keywords

  • Arabic
  • Database
  • handwritten
  • text recognition

ASJC Scopus subject areas

  • Signal Processing
  • Linguistics and Language
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Synthesizing versus Augmentation for Arabic Word Recognition with Convolutional Neural Networks'. Together they form a unique fingerprint.

Cite this