Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance

Wei Zhou; Alireza Ahrary; Sei Ichiro Kamata

doi:10.1109/ISCE.2009.5156971

Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance

Wei Zhou^*, Alireza Ahrary, Sei Ichiro Kamata

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Citations (Scopus)

Abstract

This paper presents a novel algorithm for face recognition. Local Quaternion Patters (LQP) is proposed for presenting the feature parts in the face. To keep the spatial feature of the face, an asymmetric similarity measure Weighted Spatially constrained Earth Mover's Distance (WSEMD) is studied for classification. In this step, the source image is partitioned into non overlapping local patches while the destination image is represented as a set of overlapping local patches at different positions and Gaussian Kernel is used. Finally, local and global weighting is applied to get a more accurate classifier. To evaluate the proposed method and its performance, three well-known and challenge face databases - ORL, Yale and FERET are used in our study. The experimental results show that the proposed method has higher accuracy than some other classic methods.

Original language	English
Title of host publication	2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009
Pages	285-289
Number of pages	5
DOIs	https://doi.org/10.1109/ISCE.2009.5156971
Publication status	Published - 2009 Oct 27
Event	2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009 - Kyoto, Japan Duration: 2009 May 25 → 2009 May 28

Publication series

Name	Digest of Technical Papers - IEEE International Conference on Consumer Electronics
ISSN (Print)	0747-668X

Conference

Conference	2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009
Country/Territory	Japan
City	Kyoto
Period	09/5/25 → 09/5/28

Keywords

Face recognition
Feature extraction
Local quaternion patterns (LQP)
WSEMD

ASJC Scopus subject areas

Industrial and Manufacturing Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/ISCE.2009.5156971

Cite this

Zhou, W., Ahrary, A., & Kamata, S. I. (2009). Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance. In 2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009 (pp. 285-289). Article 5156971 (Digest of Technical Papers - IEEE International Conference on Consumer Electronics). https://doi.org/10.1109/ISCE.2009.5156971

Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance. / Zhou, Wei; Ahrary, Alireza; Kamata, Sei Ichiro.
2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009. 2009. p. 285-289 5156971 (Digest of Technical Papers - IEEE International Conference on Consumer Electronics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zhou, W, Ahrary, A & Kamata, SI 2009, Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance. in 2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009., 5156971, Digest of Technical Papers - IEEE International Conference on Consumer Electronics, pp. 285-289, 2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009, Kyoto, Japan, 09/5/25. https://doi.org/10.1109/ISCE.2009.5156971

@inproceedings{17305d5a2baa4779bc44df7782240b68,

title = "Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance",

abstract = "This paper presents a novel algorithm for face recognition. Local Quaternion Patters (LQP) is proposed for presenting the feature parts in the face. To keep the spatial feature of the face, an asymmetric similarity measure Weighted Spatially constrained Earth Mover's Distance (WSEMD) is studied for classification. In this step, the source image is partitioned into non overlapping local patches while the destination image is represented as a set of overlapping local patches at different positions and Gaussian Kernel is used. Finally, local and global weighting is applied to get a more accurate classifier. To evaluate the proposed method and its performance, three well-known and challenge face databases - ORL, Yale and FERET are used in our study. The experimental results show that the proposed method has higher accuracy than some other classic methods.",

keywords = "Face recognition, Feature extraction, Local quaternion patterns (LQP), WSEMD",

author = "Wei Zhou and Alireza Ahrary and Kamata, {Sei Ichiro}",

year = "2009",

month = oct,

day = "27",

doi = "10.1109/ISCE.2009.5156971",

language = "English",

isbn = "9781424429769",

series = "Digest of Technical Papers - IEEE International Conference on Consumer Electronics",

pages = "285--289",

booktitle = "2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009",

note = "2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009 ; Conference date: 25-05-2009 Through 28-05-2009",

}

TY - GEN

T1 - Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance

AU - Zhou, Wei

AU - Ahrary, Alireza

AU - Kamata, Sei Ichiro

PY - 2009/10/27

Y1 - 2009/10/27

N2 - This paper presents a novel algorithm for face recognition. Local Quaternion Patters (LQP) is proposed for presenting the feature parts in the face. To keep the spatial feature of the face, an asymmetric similarity measure Weighted Spatially constrained Earth Mover's Distance (WSEMD) is studied for classification. In this step, the source image is partitioned into non overlapping local patches while the destination image is represented as a set of overlapping local patches at different positions and Gaussian Kernel is used. Finally, local and global weighting is applied to get a more accurate classifier. To evaluate the proposed method and its performance, three well-known and challenge face databases - ORL, Yale and FERET are used in our study. The experimental results show that the proposed method has higher accuracy than some other classic methods.

AB - This paper presents a novel algorithm for face recognition. Local Quaternion Patters (LQP) is proposed for presenting the feature parts in the face. To keep the spatial feature of the face, an asymmetric similarity measure Weighted Spatially constrained Earth Mover's Distance (WSEMD) is studied for classification. In this step, the source image is partitioned into non overlapping local patches while the destination image is represented as a set of overlapping local patches at different positions and Gaussian Kernel is used. Finally, local and global weighting is applied to get a more accurate classifier. To evaluate the proposed method and its performance, three well-known and challenge face databases - ORL, Yale and FERET are used in our study. The experimental results show that the proposed method has higher accuracy than some other classic methods.

KW - Face recognition

KW - Feature extraction

KW - Local quaternion patterns (LQP)

KW - WSEMD

UR - http://www.scopus.com/inward/record.url?scp=70350244987&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350244987&partnerID=8YFLogxK

U2 - 10.1109/ISCE.2009.5156971

DO - 10.1109/ISCE.2009.5156971

M3 - Conference contribution

AN - SCOPUS:70350244987

SN - 9781424429769

T3 - Digest of Technical Papers - IEEE International Conference on Consumer Electronics

SP - 285

EP - 289

BT - 2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009

T2 - 2009 IEEE 13th International Symposium on Consumer Electronics, ISCE 2009

Y2 - 25 May 2009 through 28 May 2009

ER -

Face recognition using local quaternion patters and weighted spatially constrained earth mover's distance

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this