Global Minimum for Active Contour Models: A Minimal Path Approach

  • Authors:

  • Laurent D. Cohen;Ron Kimmel

  • Affiliations:

  • CEREMADE, U.R.A. CNRS 749, Université Paris IX, Dauphine, Place du Marechal de Lattre de Tassigny 75775 Paris CEDEX 16, France. E-mail: Cohen@ceremade.dauphine.fr;Mailstop 50A-2152, Lawrence Berkeley National Laboratory, University of California, Berkeley, CA 94720, USA. E-mail: Ron@math.lbl.gov

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 1997

Quantified Score

Hi-index 0.01

Visualization

Abstract

A new boundary detection approach for shape modeling is presented. Itdetects the global minimum of an active contour model‘s energybetween two end points. Initialization is made easier and the curveis not trapped at a local minimum by spurious edges. We modify the“snake” energy by including the internal regularization term in theexternal potential term. Our method is based on finding a path ofminimal length in a Riemannian metric. We then make use of a newefficient numerical method to find this shortest path.It is shown that the proposed energy, though based only on apotential integrated along the curve, imposes a regularization effectlike snakes. We explore the relation between the maximum curvaturealong the resulting contour and the potential generated from the image.The method is capable to close contours, given only one point on theobjects‘ boundary by using a topology-based saddle search routine.We show examples of our method applied to real aerial andmedical images.