Text-independent writer identification using SIFT descriptor and contour-directional feature

Yu Jie Xiong, Ying Wen, Patrick S.P. Wang, Yue Lu

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

40 Scopus citations

Abstract

This paper presents a method for text-independent writer identification using SIFT descriptor and contour-directional feature (CDF). The proposed method contains two stages. In the first stage, a codebook of local texture patterns is constructed by clustering a set of SIFT descriptors extracted from images. Using this codebook, the occurrence histograms are calculated to determine the similarities between different images. For each image, we obtain a candidate list of reference images. The next stage is to refine the candidate list using the contour-directional feature and SIFT descriptor. The proposed method is evaluated with two datasets: the ICFHR2012-Latin dataset and the ICDAR2013 dataset. Experimental results show that the proposed method outperforms the state-of-the-art algorithms and archives the best performance.

Original languageEnglish
Title of host publication13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings
PublisherIEEE Computer Society
Pages91-95
Number of pages5
ISBN (Electronic)9781479918058
DOIs
StatePublished - 20 Nov 2015
Event13th International Conference on Document Analysis and Recognition, ICDAR 2015 - Nancy, France
Duration: 23 Aug 201526 Aug 2015

Publication series

NameProceedings of the International Conference on Document Analysis and Recognition, ICDAR
Volume2015-November
ISSN (Print)1520-5363

Conference

Conference13th International Conference on Document Analysis and Recognition, ICDAR 2015
Country/TerritoryFrance
CityNancy
Period23/08/1526/08/15

Keywords

  • SIFT descriptor and contour-directional feature
  • Writer identification
  • codebook
  • text-independent

Fingerprint

Dive into the research topics of 'Text-independent writer identification using SIFT descriptor and contour-directional feature'. Together they form a unique fingerprint.

Cite this