Mateo Agulló, Carlos, Gil, Pablo, Torres, Fernando
Computation of Curvature Skeleton to Measure Deformations in Surfaces
Filipe, J., etal. (eds.). Informatics in Control, Automation and Robotics 12th International Conference, ICINCO 2015 Colmar, France, July 21-23, 2015 Revised Selected Papers, Lecture Notes in Electrical Engineering 383, pp. 197-207. doi:10.1007/978-3-319-31898-1_11
URI: http://hdl.handle.net/10045/58321
DOI: 10.1007/978-3-319-31898-1_11
ISSN: 1876-1100
ISBN: 978-3-319-31896-7
Abstract: 
This work presents a method to analyse 3D flat objects and to measure variations of its surface. The method represents of the objects using processsing techniques based on point cloud. The proposed method is focused on the deformation detection of elastic objects which are formed by flat faces. These deformations are usually caused when two bodies, a solid and another elastic object, come in contact and there are contact pressures among their faces. Our method describes an algorithm to estimate the deformation shape. It is done by calculating its skeleton. Particularly, the algorithm calculates the curvature values of the surface points on the object using an analysis of eigenvectors and eigenvalues. Afterwards, the points of the similar curvature are grouped in level curvatures. Finally, a set of the minimum path among level curvatures are calculated to obtain the skeleton. The paper shows a set of experiments which simulate the deformations caused by a robot hand in manipulation tasks of flat objects.
Keywords:3D shape, Surface, Curvature, Surface normal, Geometric modelling, Deformations visual perception
Springer International Publishing Switzerland
info:eu-repo/semantics/article