Efficient Local Image Descriptors


Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription Access

Abstract

We consider descriptors in polar and log-polar coordinates that produce compact description of the image near interest points. These descriptors may be used to establish one-to-one correspondence between points in two images. Algorithms to compute these descriptors are described and the results of a numerical experiment comparing them with SIFT and SURF descriptors are reported. The numerical experiment suggests that the proposed descriptors are highly efficient for finding conjugate points in two images.

About the authors

A. A. Lukianitsa

Faculty of Computational Mathematics and Cybernetics, Moscow State University

Author for correspondence.
Email: andrei.luk@gmail.com
Russian Federation, Moscow

Supplementary files

Supplementary Files
Action
1. JATS XML

Copyright (c) 2017 Springer Science+Business Media New York