Full Text:   <996>

Summary:  <1083>

CLC number: TP391

On-line Access: 2020-09-09

Received: 2019-10-11

Revision Accepted: 2020-03-23

Crosschecked: 2020-07-23

Cited: 0

Clicked: 2309

Citations:  Bibtex RefMan EndNote GB/T7714

 ORCID:

Jian Zhang

https://orcid.org/0000-0001-5764-9351

Heng Zhang

https://orcid.org/0000-0002-4201-3892

-   Go to

Article info.
Open peer comments

Frontiers of Information Technology & Electronic Engineering  2020 Vol.21 No.9 P.1334-1345

http://doi.org/10.1631/FITEE.1900552


Subspace transform induced robust similarity measure for facial images


Author(s):  Jian Zhang, Heng Zhang, Li-ling Bo, Hong-ran Li, Shuai Xu, Dong-qing Yuan

Affiliation(s):  Department of Computer Engineering, Jiangsu Ocean University, Lianyungang 222005, China; more

Corresponding email(s):   zhangjian@jou.edu.cn, zhangheng@jou.edu.cn

Key Words:  Subspace analysis, Image similarity measure, Face recognition, Pattern recognition


Jian Zhang, Heng Zhang, Li-ling Bo, Hong-ran Li, Shuai Xu, Dong-qing Yuan. Subspace transform induced robust similarity measure for facial images[J]. Frontiers of Information Technology & Electronic Engineering, 2020, 21(9): 1334-1345.

@article{title="Subspace transform induced robust similarity measure for facial images",
author="Jian Zhang, Heng Zhang, Li-ling Bo, Hong-ran Li, Shuai Xu, Dong-qing Yuan",
journal="Frontiers of Information Technology & Electronic Engineering",
volume="21",
number="9",
pages="1334-1345",
year="2020",
publisher="Zhejiang University Press & Springer",
doi="10.1631/FITEE.1900552"
}

%0 Journal Article
%T Subspace transform induced robust similarity measure for facial images
%A Jian Zhang
%A Heng Zhang
%A Li-ling Bo
%A Hong-ran Li
%A Shuai Xu
%A Dong-qing Yuan
%J Frontiers of Information Technology & Electronic Engineering
%V 21
%N 9
%P 1334-1345
%@ 2095-9184
%D 2020
%I Zhejiang University Press & Springer
%DOI 10.1631/FITEE.1900552

TY - JOUR
T1 - Subspace transform induced robust similarity measure for facial images
A1 - Jian Zhang
A1 - Heng Zhang
A1 - Li-ling Bo
A1 - Hong-ran Li
A1 - Shuai Xu
A1 - Dong-qing Yuan
J0 - Frontiers of Information Technology & Electronic Engineering
VL - 21
IS - 9
SP - 1334
EP - 1345
%@ 2095-9184
Y1 - 2020
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/FITEE.1900552


Abstract: 
Similarity measure has long played a critical role and attracted great interest in various areas such as pattern recognition and machine perception. Nevertheless, there remains the issue of developing an efficient two-dimensional (2D) robust similarity measure method for images. Inspired by the properties of subspace, we develop an effective 2D image similarity measure technique, named transformation similarity measure (TSM), for robust face recognition. Specifically, the TSM method robustly determines the similarity between two well-aligned frontal facial images while weakening interference in the face recognition by linear transformation and singular value decomposition. We present the mathematical features and some odds to reveal the feasible and robust measure mechanism of TSM. The performance of the TSM method, combined with the nearest neighbor rule, is evaluated in face recognition under different challenges. Experimental results clearly show the advantages of the TSM method in terms of accuracy and robustness.

子空间变换诱导的稳健人脸图像相似度度量

张键1,张恒2,薄丽玲2,李宏然1,徐帅1,袁冬青2
1江苏海洋大学计算机工程学院,中国连云港市,222005
2江苏海洋大学数学系,中国连云港市,222005

摘要:相似度度量方法在许多领域(如模式识别与机器感知)扮演着重要角色,引起国内外学者重点关注。当前,为图像构建二维稳健的相似度度量方法仍是重要研究课题。本文针对稳健人脸识别问题,基于子空间性质,提出一种有效且稳健的二维图像相似度度量方法。该方法通过线性变换与奇异值分解,度量两幅对齐人脸图像的相似度,同时消弱人脸识别过程中的干扰。展示了该方法的数学特征及度量特性,进而揭示所提方法的可行性与稳健机制。结合最近邻分类器,评估了所提方法在不同挑战下的人脸识别性能。实验结果表明所提方法在准确性和稳健性方面具有一定优势。

关键词:子空间分析;图像相似度度量;人脸识别;模式识别

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

[1]Adini Y, Moses Y, Ullman S, 1997. Face recognition: the problem of compensating for changes in illumination direction. IEEE Trans Patt Anal Mach Intell, 19(7):721-732.

[2]Bowyer KW, Phillips PJ, 1998. Empirical Evaluation Techniques in Computer Vision. IEEE Computer Society Press, Washington, USA.

[3]Chen SW, Dai XL, Pan BB, et al., 2015. A novel discriminant criterion based on feature fusion strategy for face recognition. Neurocomputing, 159:67-77.

[4]Cover T, Hart P, 1967. Nearest neighbor pattern classification. IEEE Trans Inform Theory, 13(1):21-27.

[5]Demirel H, Anbarjafari G, Jahromi MNS, 2008. Image equalization based on singular value decomposition. Proc 23rd Int Symp on Computer and Information Sciences, p.1-5.

[6]Demmel J, Kahan W, 1990. Accurate singular values of bidiagonal matrices. SIAM J Sci Stat Comput, 11(5):873-912.

[7]Deng WH, Hu JN, Guo J, 2012. Extended SRC: undersampled face recognition via intraclass variant dictionary. IEEE Trans Patt Anal Mach Intell, 34(9):1864-1870.

[8]Georghiades AS, Belhumeur PN, Kriegman DJ, 2001. From few to many: illumination cone models for face recognition under variable lighting and pose. IEEE Trans Patt Anal Mach Intell, 23(6):643-660.

[9]Gross R, Matthews I, Cohn J, et al., 2010. Multi-PIE. Image Vis Comput, 28(5):807-813.

[10]Gu ZH, Shao M, Li LY, et al., 2012. Discriminative metric: Schatten norm vs. vector norm. Proc 21st Int Conf on Pattern Recognition, p.1213-1216.

[11]He XT, Peng YX, Zhao JJ, 2019. Fast fine-grained image classification via weakly supervised discriminative localization. IEEE Trans Circ Syst Video Technol, 29(5):1394-1407.

[12]Lee KC, Ho J, Kriegman DJ, 2005. Acquiring linear subspaces for face recognition under variable lighting. IEEE Trans Patt Anal Mach Intell, 27(5):684-698.

[13]Liu CJ, 2014. Discriminant analysis and similarity measure. Patt Recogn, 47(1):359-367.

[14]Liu Y, Jin R, 2006. Distance Metric Learning: a Comprehensive Survey. Michigan State University, p.4.

[15]Martínez A, Benavente R, 1998. The AR Face Database. 24 CVC Technical Report.

[16]Naseem I, Togneri R, Bennamoun M, 2010. Linear regression for face recognition. IEEE Trans Patt Anal Mach Intell, 32(11):2106-2112.

[17]Paredes R, Vidal E, 2006. Learning weighted metrics to minimize nearest-neighbor classification error. IEEE Trans Patt Anal Mach Intell, 28(7):1100-1110.

[18]Peng YX, He XT, Zhao JJ, 2017. Object-part attention model for fine-grained image classification. IEEE Trans Image Process, 27(3):1487-1500.

[19]Peng YX, Qi JW, Yuan YX, 2018. Modality-specific cross- modal similarity measurement with recurrent attention network. IEEE Trans Image Process, 27(11):5585-5599.

[20]Perlibakas V, 2004. Distance measures for PCA-based face recognition. Patt Recogn Lett, 25(6):711-724.

[21]Sebe N, Lew MS, Huijsmans DP, 2000. Toward improved ranking metrics. IEEE Trans Patt Anal Mach Intell, 22(10):1132-1143.

[22]Sun YY, Tong F, Zhang ZK, et al., 2018. Throughput modeling and analysis of random access in narrowband Internet of Things. IEEE Int Things J, 5(3):1485-1493.

[23]Vivek EP, Sudha N, 2007. Robust Hausdorff distance measure for face recognition. Patt Recogn, 40(2):431-442.

[24]Wagner A, Wright J, Ganesh A, et al., 2012. Toward a practical face recognition system: robust alignment and illumination by sparse representation. IEEE Trans Patt Anal Mach Intell, 34(2):372-386.

[25]Wang H, 2006. Nearest neighbors by neighborhood counting. IEEE Trans Patt Anal Mach Intell, 28(6):942-953.

[26]Wang XG, Tang XO, 2009. Face photo-sketch synthesis and recognition. IEEE Trans Patt Anal Mach Intell, 31(11):1955-1967.

[27]Wen Y, Zhang L, von Deneen KM, et al., 2016. Face recognition using discriminative locality preserving vectors. Dig Signal Process, 50:103-113.

[28]Wright J, Yang AY, Ganesh A, 2009. Robust face recognition via sparse representation. IEEE Trans Patt Anal Mach Intell, 31(2):210-227.

[29]Wu J, Shen H, Li YD, et al., 2013. Learning a hybrid similarity measure for image retrieval. Patt Recogn, 46(11):2927-2939.

[30]Wu MC, He SB, Zhang YT, et al., 2019. A tensor-based framework for studying eigenvector multicentrality in multilayer networks. PNAS, 116(31):15407-15413.

[31]Yambor WS, Draper BA, Beveridge JR, 2002. Analyzing PCA-based face recognition algorithms: eigenvector selection and distance measures. Empirical Evaluation Methods in Computer Vision, p.39-60.

[32]Zhang DQ, Chen SC, Zhou ZH, 2005. A new face recognition method based on SVD perturbation for single example image per person. Appl Math Comput, 163(2):895-907.

[33]Zhang J, Yang J, 2014. Linear reconstruction measure steered nearest neighbor classification framework. Patt Recogn, 47(4):1709-1720.

[34]Zhang J, Yang J, Qian JJ, 2015. Nearest orthogonal matrix representation for face recognition. Neurocomputing, 151:471-480.

[35]Zhang YM, He SB, Chen JM, 2016. Data gathering optimization by dynamic sensing and routing in rechargeable sensor networks. IEEE/ACM Trans Netw, 24(3):1632-1646.

[36]Zhou CW, Gu YJ, He SB, et al., 2018. A robust and efficient algorithm for coprime array adaptive beamforming. IEEE Trans Veh Technol, 67(2):1099-1112.

[37]Zhuang LS, Yang AY, Zhou ZH, et al., 2013. Single-sample face recognition with image corruption and misalignment via sparse illumination transfer. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.3546-3553.

[38]Zou WW, Yuen PC, 2010. Discriminability and reliability indexes: two new measures to enhance multi-image face recognition. Patt Recogn, 43(10):3483-3493.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2022 Journal of Zhejiang University-SCIENCE