A Markovian Approach for Handwritten Document Segmentation

We address in this paper the problem of segmenting complex handritten pages such as novelist drafts or authorial manuscripts. We propose to use stochastic and contextual models in order to cope with local spatial variability, and to take into account some prior knowledge about the global structure of the document image. The models we propose to use are Markov Random Field models.

• Type: Conference
• Year: 2006
• Where: ICPR
• Authors: Stéphane Nicolas, Thierry Paquet, Laurent Heutte
• Keywords: Computer Vision, Contextual Models, ICPR 2006, Local Spatial Variability, Random Field Models