Home // PATTERNS 2011, The Third International Conferences on Pervasive Patterns and Applications // View article

Word Spotting for Arabic Handwritten Historical Document Retrieval using Generalized Hough Transform

Authors:
Nabil Aouadi
Afef Kacem

Keywords: Generalized Hough Transform; word spotting; pattern recognition; image processing

Abstract:
Because of the high noise levels in historical documents and the great amount of variability in handwriting, handwritten historical documents are currently transcribed by hand. Easy access to such documents requires an index, which is currently created manually at great cost. The goal of the Word Spotting idea, applied to handwritten documents, is to greatly reduce the amount of annotation work that has to be performed, by grouping all words into clusters. This paper explores the use of GHT (Generalized Hough Transform) in case of word spotting for Arabic handwritten historical document retrieval. We applied GHT to identify all positions of a given word in a document. It has the advantage of being relatively unaffected by image noise. Experiments that have been conducted on the historical documents of the Tunisian national Archive show the advantage of the proposed approach.

Pages: 67 to 71

Copyright: Copyright (c) IARIA, 2011

Publication date: September 25, 2011

Published in: conference

ISSN: 2308-3557

ISBN: 978-1-61208-158-8

Location: Rome, Italy

Dates: from September 25, 2011 to September 30, 2011