Presentation 1997/6/20
Rigid Object Extraction for Security Services
Akinori Kosako, Naoki Mukawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method to extracting a dangerous rigid object from sensing images. First, we describe the requirements for extracting the object for security services: (1) an object should be extracted under the varing light condition, (2) the object should be extracted against variation of its pose and against occlusion, (3) it should be extracted from limitted evidence. To overcome (2) we propose a method applying Parameteric Eigenspace to small sub-image regions. It integrates local features based on Bayesian apploach, but it does not satisfy the requirement (3). In this paper we introduce Dempster-Shafer theory. The proposal method exhibits good performance for security scenes.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) object extraction / local feature / Dempster-Shafer's theory / Parametric Eigenspace Representation
Paper # PRMU97-51
Date of Issue

Conference Information
Committee PRMU
Conference Date 1997/6/20(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Rigid Object Extraction for Security Services
Sub Title (in English)
Keyword(1) object extraction
Keyword(2) local feature
Keyword(3) Dempster-Shafer's theory
Keyword(4) Parametric Eigenspace Representation
1st Author's Name Akinori Kosako
1st Author's Affiliation NTT DATA CORPORATION Laboratories for Information Technology()
2nd Author's Name Naoki Mukawa
2nd Author's Affiliation NTT DATA CORPORATION Laboratories for Information Technology
Date 1997/6/20
Paper # PRMU97-51
Volume (vol) vol.97
Number (no) 113
Page pp.pp.-
#Pages 8
Date of Issue