95 articles 

inria-00104466, version 1

Comparison of Gabor-Based Features for Writer Identification of Farsi/Arabic Handwriting

F. Shahabi () 1, M. Rahmati () 1

Tenth International Workshop on Frontiers in Handwriting Recognition (2006)

Abstract: Writer identification recently has been studied and it has a wide variety of applications. Most studies are based on English documents with the assumption that the written text is fixed (text-dependent methods) and no research has been reported on Farsi or Arabic documents. In this paper, we have proposed a method for off-line writer identification based on Farsi handwriting, which is text-independent. Based on idea that has been presented in previous studies, in this paper we assume handwriting as texture image and a set of features which are based on multi-channel Gabor filtering and co-occurrence matrix features, are extracted from preprocessed image of documents. Experimental results demonstrate that the best result, 92% correct identification in a hit list with size 3 and 88% in a hit list with size 1, is acquired by using Gaborenergy features on Farsi handwritten documents from 25 peoples.

  • 1:  Department of Computer Engineering (Sharif University of Technology)
  • Sharif University of Technology
  • Domain : Computer Science/Document and Text Processing
    Computer Science/Computer Vision and Pattern Recognition
  • Keywords : Gabor filters – handwriting – multi-channel filtering – writer identification
  • Comment : http://www.suvisoft.com
  • inria-00104466, version 1
  • oai:hal.inria.fr:inria-00104466
  • From: 
  • Submitted on: Friday, 6 October 2006 15:45:53
  • Updated on: Friday, 6 October 2006 15:49:17