Meaningful Scales Detection: an Unsupervised Noise Detection Algorithm for Digital Contours

Bertrand Kerautret (1) , Jacques-Olivier Lachaud (2)

1 ADAGIO - Applying Discrete Algorithms to Genomics and Imagery
2 LAMA - Laboratoire de Mathématiques

This work presents an algorithm which permits to detect locally on digital contours what is the amount of noise estimated from a given maximal scale. The method is based on the asymptotic properties of the length of the maximal segment primitive.

Domaines: Traitement des images [eess.IV]

Bertrand Kerautret, Jacques-Olivier Lachaud. Meaningful Scales Detection: an Unsupervised Noise Detection Algorithm for Digital Contours. Image Processing On Line, 2014, Special Issue on Discrete Geometry (DGCI 2011), 4, pp.18. ⟨10.5201/ipol.2014.75⟩. ⟨hal-01112936⟩