Curvature dependent skeletonization

Authors:
Andrea Torsello;Edwin R. Hancock
Affiliations:
Department of Computer Science, University of York, York, UK;Department of Computer Science, University of York, York, UK
Venue:
SCIA'03 Proceedings of the 13th Scandinavian conference on Image analysis
Year:
2003

Citing 6
Cited 0

Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations

Journal of Computational Physics
Ridge points in Euclidean distance maps

Pattern Recognition Letters
Hamilton-Jacobi Skeletons

International Journal of Computer Vision
Some examples of algorithm analysis in computational geometry by means of mathematical morphological techniques

Proceedings of the Workshop on Geometry and Robotics
A shock grammar for recognition

CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
View-Based 3-D Object Recognition using Shock Graphs

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 3 - Volume 3

Quantified Score

Hi-index	0.00

Visualization

Abstract

The Hamilton-Jacobi approach has proved to be a powerful and elegant method for extracting the skeleton of a shape. The approach is based on the fact that the dynamics of the inward evolving boundary is conservative everywhere except at skeletal points. Nonetheless this method appears to overlook the fact that the linear density of the evolving boundary front is not constant where the front is curved. In this paper we present an analysis which takes into account variations of density due to boundary curvature. This yields a skeletonization algorithm that is both better localized and less susceptible to boundary noise than the Hamilton-Jacobi method.