Fast Unsupervised Segmentation Using Active Contours and Belief Functions

Foued Derraz , Laurent Peyrodie (1) , Abdelmalik Taleb-Ahmed (2) , Miloud Boussahla , Gérard Forzy

1 LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
2 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201

In this paper, we study Active Contours AC based globally segmentation for vector valued images using evidential Kullback-Leibler KL distance. We investigate the evidential framework to fuse multiple features issued from vector-valued images. This formulation has two main advantages: 1 by the combination of foreground/background issued from the multiple channels in the same framework. 2 the incorporation of the heterogeneous knowledge and the reduction of the imprecision due to the noise. The statistical relation between the image channels is ensured by the Dempster-Shafer rule. We illustrate the performance of our segmentation algorithm using some challenging color and textured images.

Mots clés: Active Contours; Characteristic function; Evidential Kullback-Leibler distance; Belief Functions; Dempster-Shafer rule

Domaines: Sciences de l'ingénieur [physics]

Foued Derraz, Laurent Peyrodie, Abdelmalik Taleb-Ahmed, Miloud Boussahla, Gérard Forzy. Fast Unsupervised Segmentation Using Active Contours and Belief Functions. 15th International Conference, CAIP 2013, Aug 2013, York, United Kingdom. pp.278-285, ⟨10.1007/978-3-642-40261-6_33⟩. ⟨hal-03614737⟩