Presentation 1999/10/21
A Feature based on effective divisions for Personal Identification
Yasuhiro TANAKA, Yoshito MEKADA, Masao KASUGA, Hiroharu YOSHIKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, it has become necessary to devalop a facial discrimination system. We have developed a feature that corresponds to personal facial characteristic by using the normal vectors of approximated plane of small areas. In this paper, we improve this feature by extracting the effctive divisions for personal identification based on our features.The result showed that the accuracy of personal identification was progressed using improved feature.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Personal identification / Image processing / Average cosine / Effective divisions / Feature
Paper # HCS99-50
Date of Issue

Conference Information
Committee HCS
Conference Date 1999/10/21(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 Human Communication Science (HCS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Feature based on effective divisions for Personal Identification
Sub Title (in English)
Keyword(1) Personal identification
Keyword(2) Image processing
Keyword(3) Average cosine
Keyword(4) Effective divisions
Keyword(5) Feature
1st Author's Name Yasuhiro TANAKA
1st Author's Affiliation Department of Information Sience, Faculty of Engineering Utsunomiya University()
2nd Author's Name Yoshito MEKADA
2nd Author's Affiliation Department of Information Sience, Faculty of Engineering Utsunomiya University
3rd Author's Name Masao KASUGA
3rd Author's Affiliation Department of Information Sience, Faculty of Engineering Utsunomiya University
4th Author's Name Hiroharu YOSHIKAWA
4th Author's Affiliation Ricoh system kaihatsu Company, Ltd.
Date 1999/10/21
Paper # HCS99-50
Volume (vol) vol.99
Number (no) 384
Page pp.pp.-
#Pages 6
Date of Issue