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


This work presents an algorithm which permits to detect locally on digital contour 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.

IPOL Journal, 4: 98-115, 2014
Jacques-Olivier Lachaud
Jacques-Olivier Lachaud
Professor of Computer Science

My research interests include digital geometry, geometry processing, image analysis, variational models and discrete calculus.