Title:
分類方法と装置
Document Type and Number:
Japanese Patent JP4723724
Kind Code:
B2
Abstract:
A method of deriving a classification for classifying items of currency into two or more classes comprises measuring known samples for each class, selecting a function corresponding to a non-linear mapping of the feature vector space to a second higher-dimensional space, mapping feature vectors to image vectors, and deriving coefficients representing N-1 axes, where N is the number of classes, in the second space, obtaining values representing the projections of the image vectors for the measured samples onto the N-1 axes, and using those values to derive a separating function for separating the classes equivalent to a function in the second space.
More Like This:
Inventors:
Bourd, Gaston
Application Number:
JP2000585834A
Publication Date:
July 13, 2011
Filing Date:
December 01, 1999
Export Citation:
Assignee:
MI Incorporated
International Classes:
G07D7/00; G06K9/62; G07D5/00
Other References:
Christopher M. Bishop,Neural Networks for Pattern Recognition,英国,Oxford Univ Pr,1996年 1月18日,105-111
BURGES C J C,Simplified Support Vector Decision Rules,MACHINE LEARNING,PROCEEDINGS OF THE INTERNATIONAL CONFERENCE,1996年 7月 3日,P71-77
BURGES C J C,Simplified Support Vector Decision Rules,MACHINE LEARNING,PROCEEDINGS OF THE INTERNATIONAL CONFERENCE,1996年 7月 3日,P71-77
Attorney, Agent or Firm:
Okabe
Masao Okabe
Nobuaki Kato
Shinichi Usui
Takao Ochi
Teruhisa Motomiya
Asahi Shinmitsu
Katsumi Miyama
Masao Okabe
Nobuaki Kato
Shinichi Usui
Takao Ochi
Teruhisa Motomiya
Asahi Shinmitsu
Katsumi Miyama