Graph-theoretic image alignment using topological features

Authors:
Waleed Mohamed;A. Ben Hamza;Khaled Gharaibeh
Affiliations:
Concordia Institute for Information Systems Engineering, Concordia University, Montréal, QC, Canada;Concordia Institute for Information Systems Engineering, Concordia University, Montréal, QC, Canada;Telecommunication Engineering Department, Hijawi Faculty of Engineering Technology, Yarmouk University, Irbid, Jordan
Venue:
CompIMAGE'10 Proceedings of the Second international conference on Computational Modeling of Objects Represented in Images
Year:
2010

Citing 3
Cited 0

Alignment by Maximization of Mutual Information

International Journal of Computer Vision
A generalized divergence measure for robust image registration

IEEE Transactions on Signal Processing
Using Spanning Graphs for Efficient Image Registration

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we introduce a feature-based image alignment method using topological singularities. The main idea behind our proposed framework is to encode a medical image into a set of Morse critical points. Then an entropic dissimilarity measure between the Morse features of the target and the reference images is maximized to bring the data into alignment. We also show that maximizing this divergence measure leads to minimizing the total length of the joint minimal spanning tree between the features of the misaligned medical images. Illustrative experimental results clearly show the much improved performance and the registration accuracy of the proposed technique.