Using 2D Topological Map Information in a Markovian Image Segmentation

Guillaume Damiand (1) , Olivier Alata (1) , Camille Bihoreau (1)

1 SIC - SIGNAL-IMAGE-COMMUNICATION

Topological map is a mathematical model of labeled image representation which contains both topological and geometrical information. In this work, we use this model to improve a Markovian seg-mentation algorithm. Image segmentation methods based on Markovian assumption consist in optimizing a Gibbs energy function. This energy function can be given by a sum of potentials which could be based on the shape or the size of a region, the number of adjacencies,.. . and can be computed by using topological map. In this work we propose the integration of a new potential: the global linearity of the boundaries, and show how this potential can be extracted from the topological map. Moreover, to decrease the complexity of our algorithm, we propose a local modification of the topological map in order to avoid the reconstruction of the entire structure.

Mots clés: Markovian segmentation; topological maps; region segmentation; boundaries linearity

Domaines: Traitement des images [eess.IV]

Guillaume Damiand, Olivier Alata, Camille Bihoreau. Using 2D Topological Map Information in a Markovian Image Segmentation. 11th International Conference on Discrete Geometry for Computer Imagery (DGCI 2003), Nov 2003, Naples, Italy. pp.288-297, ⟨10.1007/978-3-540-39966-7_27⟩. ⟨hal-01513073⟩