Information Extraction from Old Images of Documents for Indexing

Mickaël Coustaty (1) , Sloven Dubois (1, 2) , Jean-Marc Ogier (1) , Michel Menard (1)

1 L3I - Laboratoire Informatique, Image et Interaction - EA 2118
2 MIA - Mathématiques, Image et Applications - EA 3165

This paper present a new method to extract areas of interest in drop caps and particularly the most important shape: Letter itself. This method relies on a combination of a Aujol and Chambolle algorithm and a Segmentation using a Zipf Law and can be enhanced as a three-step process: 1)Decomposition in layers 2)Segmentation using a Zipf Law 3)Selection of the connected components.

Mots clés: Zipf Law; Decomposition; Connected Components; Information Extraction

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Mickaël Coustaty, Sloven Dubois, Jean-Marc Ogier, Michel Menard. Information Extraction from Old Images of Documents for Indexing. Eighth International Workshop on Graphics Recognition - GREC 2009, Jul 2009, La Rochelle, France. pp.303-307. ⟨hal-00382082⟩