Multimodal image matching via dualcodebook-based self-similarity hypercube feature descriptor and voting strategy

H. Wang, D. K. Han, Hanseok Ko

Research output: Contribution to journalArticle

Abstract

An effective feature descriptor is proposed for multimodal local-image patch matching. The conventional self-similarity hypercube (SSH) fails in multimodal image matching due to different intensities of multimodal images. To mitigate this problem, a dual-codebook clustering is proposed for generating the descriptors. It is based on extracting a codebook, respectively, from visible and thermal images but sharing the same k-means clustering index of the local features of visible and thermal image patches. The experimental results show that the proposed approach effectively solves the multimodal image quantisation problem. Moreover, a voting strategy based on the proposed similarity family function facilitates the multimodal image matching more robustly compared with the conventional state-of-the-art methods.

Original languageEnglish
Pages (from-to)1518-1520
Number of pages3
JournalElectronics Letters
Volume50
Issue number21
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Image matching
Hot Temperature

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Multimodal image matching via dualcodebook-based self-similarity hypercube feature descriptor and voting strategy. / Wang, H.; Han, D. K.; Ko, Hanseok.

In: Electronics Letters, Vol. 50, No. 21, 01.01.2014, p. 1518-1520.

Research output: Contribution to journalArticle

@article{7105d3a679da42319330ed2b590c83e6,
title = "Multimodal image matching via dualcodebook-based self-similarity hypercube feature descriptor and voting strategy",
abstract = "An effective feature descriptor is proposed for multimodal local-image patch matching. The conventional self-similarity hypercube (SSH) fails in multimodal image matching due to different intensities of multimodal images. To mitigate this problem, a dual-codebook clustering is proposed for generating the descriptors. It is based on extracting a codebook, respectively, from visible and thermal images but sharing the same k-means clustering index of the local features of visible and thermal image patches. The experimental results show that the proposed approach effectively solves the multimodal image quantisation problem. Moreover, a voting strategy based on the proposed similarity family function facilitates the multimodal image matching more robustly compared with the conventional state-of-the-art methods.",
author = "H. Wang and Han, {D. K.} and Hanseok Ko",
year = "2014",
month = "1",
day = "1",
doi = "10.1049/el.2014.1802",
language = "English",
volume = "50",
pages = "1518--1520",
journal = "Electronics Letters",
issn = "0013-5194",
publisher = "Institution of Engineering and Technology",
number = "21",

}

TY - JOUR

T1 - Multimodal image matching via dualcodebook-based self-similarity hypercube feature descriptor and voting strategy

AU - Wang, H.

AU - Han, D. K.

AU - Ko, Hanseok

PY - 2014/1/1

Y1 - 2014/1/1

N2 - An effective feature descriptor is proposed for multimodal local-image patch matching. The conventional self-similarity hypercube (SSH) fails in multimodal image matching due to different intensities of multimodal images. To mitigate this problem, a dual-codebook clustering is proposed for generating the descriptors. It is based on extracting a codebook, respectively, from visible and thermal images but sharing the same k-means clustering index of the local features of visible and thermal image patches. The experimental results show that the proposed approach effectively solves the multimodal image quantisation problem. Moreover, a voting strategy based on the proposed similarity family function facilitates the multimodal image matching more robustly compared with the conventional state-of-the-art methods.

AB - An effective feature descriptor is proposed for multimodal local-image patch matching. The conventional self-similarity hypercube (SSH) fails in multimodal image matching due to different intensities of multimodal images. To mitigate this problem, a dual-codebook clustering is proposed for generating the descriptors. It is based on extracting a codebook, respectively, from visible and thermal images but sharing the same k-means clustering index of the local features of visible and thermal image patches. The experimental results show that the proposed approach effectively solves the multimodal image quantisation problem. Moreover, a voting strategy based on the proposed similarity family function facilitates the multimodal image matching more robustly compared with the conventional state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=84907887747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907887747&partnerID=8YFLogxK

U2 - 10.1049/el.2014.1802

DO - 10.1049/el.2014.1802

M3 - Article

AN - SCOPUS:84907887747

VL - 50

SP - 1518

EP - 1520

JO - Electronics Letters

JF - Electronics Letters

SN - 0013-5194

IS - 21

ER -