Significance tests and statistical inequalities for segmentation by region growing on graph

Guillaume Née (1, 2) , Stéphanie Jehan-Besson (3) , Luc Brun (2) , Marinette Revenu (2)

1 GE Healthcare - General Electric Medical Systems [Buc]
2 Equipe Image - Laboratoire GREYC - UMR6072
3 LIMOS - Laboratoire d'Informatique, de Modélisation et d'optimisation des Systèmes

Bottom-up segmentation methods merge similar neighboring regions according to a decision rule and a merging order. In this paper, we propose a contribution for each of these two points. Firstly, under statistical hypothesis of similarity, we provide an improved decision rule for region merging based on significance tests and the recent statistical inequality of McDiarmid. Secondly, we propose a dynamic merging order based on our merging predicate. This last heuristic is justified by considering an energy minimisation framework. Experimental results on both natural and medical images show the validity of our method.

Domaines: Traitement des images [eess.IV]

Guillaume Née, Stéphanie Jehan-Besson, Luc Brun, Marinette Revenu. Significance tests and statistical inequalities for segmentation by region growing on graph. International Conference on Computer Analysis of Images and Patterns (CAIP'09), 2009, Münster, Germany. pp.939-946, ⟨10.1007/978-3-642-03767-2_114⟩. ⟨hal-00812714⟩