Universidad Complutense de Madrid
E-Prints Complutense

On combining support vector machines and simulated annealing in stereovision matching

Impacto

Downloads

Downloads per month over past year

Pajares Martinsanz, Gonzalo and Cruz García, Jesús Manuel de la (2004) On combining support vector machines and simulated annealing in stereovision matching. IEEE Transactions on Systems Man and Cybernetics Part B-Cybernetics, 34 (4). pp. 1646-1657. ISSN 1083-4419

[img] PDF
Restringido a Repository staff only hasta 31 December 2020.

455kB

Official URL: http://dx.doi.org/10.1109/TSMCB.2004.827391


URLURL Type
http://ieeexplore.ieee.org/Publisher


Abstract

This paper outlines a method for solving the stereovision matching problem using edge segments as the primitives. In stereovision matching, the following constraints are commonly used: epipolar, similarity, smoothness, ordering, and uniqueness. We propose a new strategy in which such constraints are sequentially combined. The goal is to achieve high performance in terms of correct matches by combining several strategies. The contributions of this paper are reflected in the development of a similarity measure through a support vector machines classification approach; the transformation of the smoothness, ordering and epipolar constraints into the form of an energy function, through an optimization simulated annealing approach, whose minimum value corresponds to a good matching solution and by introducing specific conditions to overcome the violation of the smoothness and ordering constraints. The performance of the proposed method is illustrated by comparative analysis against some recent global matching methods.


Item Type:Article
Additional Information:

© 2004 IEEE. The authors wish to acknowledge the constructive recommendations provided by the reviewers.

Uncontrolled Keywords:Vision, Relaxation, Algorithm, Images, Optimization, Window
Subjects:Sciences > Computer science
ID Code:25395
Deposited On:23 May 2014 07:37
Last Modified:23 May 2014 07:37

Origin of downloads

Repository Staff Only: item control page