Semi-Hard Margin Support Vector Machines for Personal Authentication with an Aerial Signature Motion

被引：0

作者：

Yoshida, Takeshi ^{[1
]}

Kitamura, Takuya ^{[2
]}

机构：

[1] Univ Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki, Japan

[2] Toyama Coll, Natl Inst Technol, 13 Hongo Machi, Toyama, Japan

来源：

ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2021, PT IV | 2021年 / 12894卷

关键词：

Personal authentication; Spoofing attacks; Support vector machines;

D O I：

10.1007/978-3-030-86380-7_27

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The classifiers for personal authentication needs to be robust against spoofing attacks. In this paper, we propose semi-hard margin support vector machines (SH-SVMs) for personal authentication problems with an aerial signature motion. To be robust against spoofing attacks, SH-SVMs are trained so that false acceptance rate (FAR) of training data is zero. There are three types of SH-SVMs as follows: (1) standard SHSVM based on L2-SVM, (2) semi-hard twin SVM (SH-TWSVM) based on TWSVM, (3) semi-hard support vector data description (SH-SVDD) based SVDD. In computer experiments, we compare our methods with the conventional methods, using the aerial signature dataset and show effectiveness of SH-SVMs.

引用

页码：333 / 344

页数：12

共 50 条

[1] Support Vector Machines with the Ramp Loss and the Hard Margin Loss
Brooks, J. Paul
[J]. OPERATIONS RESEARCH, 2011, 59 (02) : 467 - 479
[2] Phase transition in the hard-margin support vector machines
Sifaou, Houssem
Kammoun, Abla
Alouini, Mohamed-Slim
[J]. 2019 IEEE 8TH INTERNATIONAL WORKSHOP ON COMPUTATIONAL ADVANCES IN MULTI-SENSOR ADAPTIVE PROCESSING (CAMSAP 2019), 2019, : 415 - 419
[3] Support vector machines for face authentication
Jonsson, K
Kittler, J
Li, YP
Matas, J
[J]. IMAGE AND VISION COMPUTING, 2002, 20 (5-6) : 369 - 375
[4] Training hard-margin support vector machines using greedy stagewise algorithm
Bo, Liefeng
Wang, Ling
Jiao, Licheng
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2008, 19 (08): : 1446 - 1455
[5] Robust relative margin support vector machines
Song Y.
Zhu W.
Xiao Y.
Zhong P.
[J]. Zhu, Wenxin (zhuwenxinyan@163.com), 1600, SAGE Publications Inc. (11): : 186 - 191
[6] Total margin algorithms in support vector machines
Yoon, M
Yun, YB
Nakayama, H
[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2004, E87D (05): : 1223 - 1230
[7] Adaptive margin support vector machines for classification
Herbrich, R
Weston, J
[J]. NINTH INTERNATIONAL CONFERENCE ON ARTIFICIAL NEURAL NETWORKS (ICANN99), VOLS 1 AND 2, 1999, (470): : 880 - 885
[8] A role of total margin in Support Vector Machines
Yoon, M
Yun, Y
Nakayama, H
[J]. PROCEEDINGS OF THE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS 2003, VOLS 1-4, 2003, : 2049 - 2053
[9] Multimodal Authentication using Qualitative Support Vector Machines
Alsaade, F.
Ariyaeeinia, A.
Meng, L.
Malegaonkar, A.
[J]. INTERSPEECH 2006 AND 9TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, VOLS 1-5, 2006, : 2454 - 2457
[10] Density-induced margin support vector machines
Zhang, Li
Zhou, Wei-Da
[J]. PATTERN RECOGNITION, 2011, 44 (07) : 1448 - 1460

← 1 2 3 4 5 →