Skip to Main content Skip to Navigation
Journal articles

Using uncertainty to link edge detection and local surface modelling

Olivier Monga 1 Nicholas Ayache 1, * Peter Sander 1 
* Corresponding author
Abstract : We establish a theoretical link between the 3D edge detection and the local surface approximation using uncertainty. As a practical application of the theory, we present a method for computing typical curvature features from 3D medical images. We use the uncertainties inherent in edge (and surface) detection in 2- and 3-dimensional images determined by quantitatively analyzing the uncertainty in edge position, orientation and magnitude produced by the multidimensional (2-D and 3-D) versions of the Monga-Deriche-Canny recursive separable edge-detector. These uncertainties allow to compute local geometric models (quadric surface patches) of the surface, which are suitable for reliably estimating local surface characteristics, for example, Gaussian and Mean curvature. We demonstrate the effectiveness of our methods compared to previous techniques. These curvatures are then used to obtain more structured features such as curvature extrema and lines of curvature extrema. The final goal is to extract robust geometric features on which registration and/or tracking procedures can rely.
Complete list of metadata
Contributor : Project-Team Asclepios Connect in order to contact the contributor
Submitted on : Friday, August 19, 2011 - 4:44:51 PM
Last modification on : Friday, February 4, 2022 - 3:22:08 AM
Long-term archiving on: : Friday, November 25, 2011 - 11:36:32 AM


Files produced by the author(s)




Olivier Monga, Nicholas Ayache, Peter Sander. Using uncertainty to link edge detection and local surface modelling. Lecture Notes in Computer Science, Springer, 1992, 511, pp.273-284. ⟨10.1007/BFb0033759⟩. ⟨inria-00615541⟩



Record views


Files downloads