Handwritten Document Segmentation Using Hidden Markov Random Fields

Stéphane Nicolas (1) , Yousri Kessentini (1) , Thierry Paquet (1) , Laurent Heutte (1)

1 DocApp - LITIS - Equipe Apprentissage

In this paper we present a method based on Hidden Markov Random Fields and 2D dynamic programming image decoding, for segmenting pages of complex handwritten manuscripts such as novelist drafts. After a formal description of the theoretical framework and the principles of the decoding method, we describe the implementation of the model and the decoding method. Then we discuss the results obtained with this approach on the drafts of the French novelist Gustave Flaubert.

Mots clés: document image analysis; handwritten documents; authorial manuscripts; page segmentation; layout analysis; image labeling; stochastic models; contextual models; markovian models; Markov random Fields; 2D dynamic programming image decoding

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Stéphane Nicolas, Yousri Kessentini, Thierry Paquet, Laurent Heutte. Handwritten Document Segmentation Using Hidden Markov Random Fields. 8th International Conference on Document Analysis and Recognition, ICDAR'2005, Aug 2005, Séoul, South Korea. pp: 212-216. ⟨hal-00509212⟩