A Triangle Division Based Point Matching for Image Registration

DOI

10.2991/icsem.2013.142How to use a DOI?

Keywords

Image Registration, Feature Matching, Delaunay Based Triangulation, Topology based Matching

Abstract

One basic requirement of image registration is the high precision of point matching. In this paper, we present a simple and robust method to get more accurate matching points by using the structural information of divided triangles. The method initially construct triangles using Delaunay Triangulation which considering rough matched points as vertices, and then iteratively removes outliers until obtain the consensus topology structure. An algorithm called K-Nearest Neighbor Dis-ratio is also presented to avoiding mismatches. Compared with RANSAC and GTM, Our proposed method acquires more inliers and demonstrates higher matching accuracy.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jian Chen
AU  - Dan Zeng
AU  - Chen Kang
AU  - Zhijiang Zhang
PY  - 2013/04
DA  - 2013/04
TI  - A Triangle Division Based Point Matching for Image Registration
BT  - Proceedings of the 2nd International Conference On Systems Engineering and Modeling (ICSEM 2013)
PB  - Atlantis Press
SP  - 697
EP  - 701
SN  - 1951-6851
UR  - https://doi.org/10.2991/icsem.2013.142
DO  - 10.2991/icsem.2013.142
ID  - Chen2013/04
ER  -