Presentation 2016-02-22
Secure classification based on kernel method using unitary transformation
Ibuki Nakamura, Yuko Saito, Sayaka Shiota, Hitoshi Kiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This study considers a template protection scheme based on an random unitary transformation, where the template consists of the features extracted from the biometric trait. In this paper, we apply the template protection scheme based on an unitary transformation to a kernel-based classification, and the recognition performance is theoretically considered. Previous studies showed that the Euclidean distance between templates protected by a unitary transform is the same as that between original ones, and there is no degradation of the recognition performances of using l2-norm minimization. In this study, we show a new property that there is no degradation of the recognition performances of using a kernel-based classification. The proposed method can obtain the same recognition performance as authentication with original templates even though protected templates are applied to the same procedure as the authentication with original templates. We perform some face recognition experiments of using a kernel eigenface as one of the kernel-based classification schemes experimentally to confirm the validity of the theory.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) unitary transform / kernel method / kernel eigenface / template protection method / biometrics
Paper # ITS2015-59,IE2015-101
Date of Issue 2016-02-15 (ITS, IE)

Conference Information
Committee ITS / IE / ITE-AIT / ITE-HI / ITE-ME / ITE-MMS / ITE-CE
Conference Date 2016/2/22(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Hokkaido Univ.
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Tomotaka Nagaosa(Kanto Gakuin Univ.) / Seishi Takamura(NTT) / Tokiichiro Takahashi(TDU) / Masayuki Sato(Univ. of Kitakyushu) / Kazuhito Murakami(Aichi Prefectural Univ.) / Eiichi Miyashita(NHK) / Koji Minami(Mitsubishi Electric Corp.)
Vice Chair Masahiro Fujii(Utsunomiya Univ.) / Tomotaka Wada(Kansai Univ.) / Takayuki Hamamoto(Tokyo Univ. of Science) / Atsuro Ichigaya(NHK) / / / Miki Haseyama(Hokkaido Univ.)
Secretary Masahiro Fujii(AIST) / Tomotaka Wada(Utsunomiya Univ.) / Takayuki Hamamoto(NTT) / Atsuro Ichigaya(Chiba Inst. of Tech.) / (Tokyo Polytechnic Univ.) / (NICT) / Miki Haseyama(NTT) / (NHK) / (NHK)
Assistant Kohei Ohno(Meiji Univ.) / Tetsuya Manabe(Saitama Univ.) / Keita Takahashi(Nagoya Univ.) / Kei Kawamura(KDDI R&D Labs.)

Paper Information
Registration To Technical Committee on Intelligent Transport Systems Technology / Technical Committee on Image Engineering / Technical Group on Artistic Image Technology / Technical Group on Human Information / Technical Group on Media Engineering / Technical Group on Multi-media Storage / Technical Group on Consumer Electronics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Secure classification based on kernel method using unitary transformation
Sub Title (in English)
Keyword(1) unitary transform
Keyword(2) kernel method
Keyword(3) kernel eigenface
Keyword(4) template protection method
Keyword(5) biometrics
1st Author's Name Ibuki Nakamura
1st Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
2nd Author's Name Yuko Saito
2nd Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
3rd Author's Name Sayaka Shiota
3rd Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
4th Author's Name Hitoshi Kiya
4th Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
Date 2016-02-22
Paper # ITS2015-59,IE2015-101
Volume (vol) vol.115
Number (no) ITS-458,IE-459
Page pp.pp.17-22(ITS), pp.17-22(IE),
#Pages 6
Date of Issue 2016-02-15 (ITS, IE)