Markov Random Field Models to Extract The Layout of Complex Handwritten Documents

We consider in this paper the problem of complex handwritten page segmentation 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. Using this model, the segmentation is performed using optimization techniques. Using the MRF framework, the segmentation is equivalent to an image labeling problem and is performed using optimization techniques.