The 'Two Steps' Skeleton - A New Morphological Algorithm for Image Representation

摘要

This paper addresses the representation of binary images by means of mathematical morphology, a relatively new nonlinear theory for image processing, based on set theory. It considers images as sets (instead of vectors, as in the classical linear image processing), which permits geometry-oriented transformations of the images. The new image representation described in this work, called "two steps" skeleton representation, is a natural extension of the morphological skeleton. This article will present the theoretical background of the morphological image representation, deduce the new representation and show some application examples. In the end, the possibility of generalizing the "two steps" skeleton representation for multi-dimensional images will be analyzed, to extend the scope of its algebraic characteristics as much as possible. The applications of the "two steps" skeleton representation that are presented will be illustrated by computer simulations.