Presentation	2012/7/12 Locality-Sensitive Hashing With Margin Based Feature Selection For Biometric Data Authentication Makiko Konoshima,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	Biometric authentication technology is widely employed around the world for ATM at financial institution, PC access management, and so on. The usual system is adopted the authentication method using ID information recorded in a card. This paper proposes a method for locality-sensitive hashing with margin based feature selection for biometric data authentication which realizes the rapid identification from a million people without ID information. The basic concept is to generate a hash function and importance for selection that targets a much larger size than the targeted bits, and to select the targeted bits in the order of descending importance. The new algorithm was applied to biometric, speech, and image datasets and compared with other methods, and the effects for each data application were verified.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)
Paper #	Vol.2012-CSEC-58 No.42,Vol.2012-SPT-4 No.42
Date of Issue

Conference Information
Committee	ISEC
Conference Date	2012/7/12(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Information Security (ISEC)
Language	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Locality-Sensitive Hashing With Margin Based Feature Selection For Biometric Data Authentication
Sub Title (in English)
Keyword(1)
1st Author's Name	Makiko Konoshima
1st Author's Affiliation	FUJITSU LABORATORIES LTD.()
Date	2012/7/12
Paper #	Vol.2012-CSEC-58 No.42,Vol.2012-SPT-4 No.42
Volume (vol)	vol.112
Number (no)	126
Page	pp.pp.-
#Pages	7
Date of Issue