Représentation des mots manuscrits par graphe pour la recherche par similarité

Peng Wang, Véronique Eglin, Christine Largeron, Christophe Garcia, Josep Lladós, Alicia Fornés

Research output: Contribution to journalArticleResearchpeer-review

Abstract

© 2014 Lavoisier. Effective information retrieval on handwritten document images has always been a challenging task. In this paper, we propose a novel handwritten word-spotting approach based on graph representation. The presented model comprises both topological and morphological signatures of handwriting. Skeleton-based graphs with the Shape Context labeled vertexes are established for connected components. Each word image is represented as a sequence of graphs. In order to be robust to the handwriting variations, an exhaustive merging process based on DTW alignment results introduced in the similarity measure between word images. With respect to the computation complexity, an approximate graph edit distance approach using bipartite matching is employed for graph matching. The experiments on the George Washington dataset and the marriage records from the Barcelona Cathedral dataset demonstrate that the proposed approach outperforms the state-of-the-art structural methods.
Original languageEnglish
Pages (from-to)53-75
JournalDocument Numerique
Volume17
Issue number3
DOIs
Publication statusPublished - 1 Jan 2014

Keywords

  • Block merging
  • DTW
  • Graph edit distance
  • Graph-based representation
  • Query by example
  • Shape context description
  • Word-spotting

Fingerprint Dive into the research topics of 'Représentation des mots manuscrits par graphe pour la recherche par similarité'. Together they form a unique fingerprint.

Cite this