Publication Details
Overview
 
 
Bart Truyen, Ioannis Pratikakis, Yang Xin, Jan Cornelis
 

Chapter in Book/ Report/ Conference proceeding

Abstract 

Matching analogous contours is of central importance in computer vision. In this paper, a new hierarchical finite difference method to match analogous contours is presented. In deriving this method, we have been inspired by a method due to Duncan2, who proposed a scheme for matching two contours based on the minimisation of a quadratic fitting criterion. This criterion consists of a curvature dependent bending energy term and a smoothness term. The innovation of our work is in the incorporation of a new smoothness term in the fitting criterion. As a direct consequence, the computational complexity is reduced and the equation corresponding to the minimisation of the fitting criterion is attributed a simple interpretation. The solution method itself is based on a standard finite difference approach. To improve the convergence of the matching process, a new multi-smoothing scheme is proposed. Experimental validation is carried out on three medical applications: (i) Matching of left ventricular contours in successive images of a time sequence of spin echo cardiac magnetic resonance (MR) images, (ii) Matching of brain object contours in consecutive slices of a digital brain atlas, (iii) Matching of brain object contours in segmented MR images to the outlines of the corresponding brain objects in a digital anatomical atlas.

Reference