Title:
人工視覚システムトレーニング方法、連想型ネットワークトレーニング方法、人工視覚システムリンクベクトルトレーニング装置、人工視覚システムリンク行列トレーニング装置、連想型ネットワークトレーニング装置及びコンピュータプログラム製品
Document Type and Number:
Japanese Patent JP4044436
Kind Code:
B2
Abstract:
An artificial vision system traning method starts by forming (S 2 ) a feature matrix including feature sample vectors and a corresponding response sample vector including response sample scalars. The method then uses an iterative procedure to determine a linkage vector linking the response sample vector to the feature matrix. This iterative procedure includes the steps: determining (S 4 ) a response sample vector error estimate in the response sample vector domain; transforming (S 6 ) the response sample vector error estimate into a corresponding linkage vector error estimate in the linkage vector domain; determining (S 7 ) a linkage vector estimate in the linkage vector domain by using the linkage vector error estimate; transforming (S 8 ) the linkage vector estimate into a corresponding response sample vector estimate in the response sample vector domain. These steps are repeated until (S 5 ) the response sample vector error estimate is sufficiently small.
Inventors:
Yosta Granlund
Application Number:
JP2002539981A
Publication Date:
February 06, 2008
Filing Date:
October 19, 2001
Export Citation:
Assignee:
Yosta Granlund
International Classes:
G06F17/16; G06N20/00; G06T7/00; G06N3/00
Domestic Patent References:
| JP2002509303A |
Foreign References:
| US5249259 | ||||
| WO2000055790A1 | ||||
| WO2000058914A1 |
Attorney, Agent or Firm:
Naoki Yoshimura