Image stitching using chaos-inspired dissimilarity measure

Taeyup Song, Changwon Jeon, Hanseok Ko

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

A novel image-stitching algorithm is proposed that uses the chaosinspired dissimilarity measure. Previous approaches to image stitching generally employed invariant local features and random sample consensus (RANSAC), to verify image matches. In panoramic stitching, however, RANSAC is highly sensitive to illumination difference between the pair of images stemming from two different exposures, which essentially results in mismatched pairs, leading to poor stitching performance. By applying chaos theory, the contextual change of an image becomes chaos-like, and results in a complex fractal trajectory in phase space. The specific chaos behaviour between a pair of images is exploited, whereby the resulting phase space exhibits strong invariance to illumination changes. Accordingly, the chaosinspired dissimilarity between images is measured, thereby the matched pairs are robustly found under the presence of illumination difference. Representative performance experiments demonstrate significant improvements over the conventional method.

Original languageEnglish
Pages (from-to)232-234
Number of pages3
JournalElectronics Letters
Volume51
Issue number3
DOIs
Publication statusPublished - 2015 Feb 5

Fingerprint

Chaos theory
Lighting
Invariance
Fractals
Trajectories
Experiments

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Image stitching using chaos-inspired dissimilarity measure. / Song, Taeyup; Jeon, Changwon; Ko, Hanseok.

In: Electronics Letters, Vol. 51, No. 3, 05.02.2015, p. 232-234.

Research output: Contribution to journalArticle

Song, Taeyup ; Jeon, Changwon ; Ko, Hanseok. / Image stitching using chaos-inspired dissimilarity measure. In: Electronics Letters. 2015 ; Vol. 51, No. 3. pp. 232-234.
@article{c9105187c2ea49dca9bbc7c845111f2a,
title = "Image stitching using chaos-inspired dissimilarity measure",
abstract = "A novel image-stitching algorithm is proposed that uses the chaosinspired dissimilarity measure. Previous approaches to image stitching generally employed invariant local features and random sample consensus (RANSAC), to verify image matches. In panoramic stitching, however, RANSAC is highly sensitive to illumination difference between the pair of images stemming from two different exposures, which essentially results in mismatched pairs, leading to poor stitching performance. By applying chaos theory, the contextual change of an image becomes chaos-like, and results in a complex fractal trajectory in phase space. The specific chaos behaviour between a pair of images is exploited, whereby the resulting phase space exhibits strong invariance to illumination changes. Accordingly, the chaosinspired dissimilarity between images is measured, thereby the matched pairs are robustly found under the presence of illumination difference. Representative performance experiments demonstrate significant improvements over the conventional method.",
author = "Taeyup Song and Changwon Jeon and Hanseok Ko",
year = "2015",
month = "2",
day = "5",
doi = "10.1049/el.2014.0981",
language = "English",
volume = "51",
pages = "232--234",
journal = "Electronics Letters",
issn = "0013-5194",
publisher = "Institution of Engineering and Technology",
number = "3",

}

TY - JOUR

T1 - Image stitching using chaos-inspired dissimilarity measure

AU - Song, Taeyup

AU - Jeon, Changwon

AU - Ko, Hanseok

PY - 2015/2/5

Y1 - 2015/2/5

N2 - A novel image-stitching algorithm is proposed that uses the chaosinspired dissimilarity measure. Previous approaches to image stitching generally employed invariant local features and random sample consensus (RANSAC), to verify image matches. In panoramic stitching, however, RANSAC is highly sensitive to illumination difference between the pair of images stemming from two different exposures, which essentially results in mismatched pairs, leading to poor stitching performance. By applying chaos theory, the contextual change of an image becomes chaos-like, and results in a complex fractal trajectory in phase space. The specific chaos behaviour between a pair of images is exploited, whereby the resulting phase space exhibits strong invariance to illumination changes. Accordingly, the chaosinspired dissimilarity between images is measured, thereby the matched pairs are robustly found under the presence of illumination difference. Representative performance experiments demonstrate significant improvements over the conventional method.

AB - A novel image-stitching algorithm is proposed that uses the chaosinspired dissimilarity measure. Previous approaches to image stitching generally employed invariant local features and random sample consensus (RANSAC), to verify image matches. In panoramic stitching, however, RANSAC is highly sensitive to illumination difference between the pair of images stemming from two different exposures, which essentially results in mismatched pairs, leading to poor stitching performance. By applying chaos theory, the contextual change of an image becomes chaos-like, and results in a complex fractal trajectory in phase space. The specific chaos behaviour between a pair of images is exploited, whereby the resulting phase space exhibits strong invariance to illumination changes. Accordingly, the chaosinspired dissimilarity between images is measured, thereby the matched pairs are robustly found under the presence of illumination difference. Representative performance experiments demonstrate significant improvements over the conventional method.

UR - http://www.scopus.com/inward/record.url?scp=84922463490&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922463490&partnerID=8YFLogxK

U2 - 10.1049/el.2014.0981

DO - 10.1049/el.2014.0981

M3 - Article

AN - SCOPUS:84922463490

VL - 51

SP - 232

EP - 234

JO - Electronics Letters

JF - Electronics Letters

SN - 0013-5194

IS - 3

ER -