PAPER SENDING

  • googleplus
  • facebook
  • twitter
  • linkedin
  • linkedin

REVISTA DYNA NEW TECHNOLOGIES REVISTA DYNA NEW TECHNOLOGIES

  • Skip to the menu
  • Skip to the content
  • DYNA Publishing
    • DYNA
    • DYNA Energy and Sustainability
    • DYNA Management
    • DYNA New Technologies
  • Journal
    • The Journal and its bodies
      • The Journal
      • Editorial Board
      • Advisory-Scientific Board
    • Diffusion & indexation database
    • Mission, Vision & Values
    • Collaborating with DYNA
  • Authors & Referees
    • Guidelines, rules and forms
    • Collaborating with Journal
  • Papers
    • Search Content
    • Volumes / Issues
    • Most downloaded
    • Sending papers
  • Forum
  • News
    • News New Technologies
    • Newsletter DNT
  • Advertising
    • Advertising at DYNA
    • Advertising rates
  • Contact
    • Contacting
  • Search
    • In this Journal
    • Search in DYNA journals
  • Submit
  • Sign in
    • Privacy Policy

Return to the menu

  • Homepage
  • Papers
  • Search Content

Search Content

×

 |    : /

Vote:

Results: 

5 points

 212  Votes

IMPROVEMENT OF THE NEOCOGNITRON TRAINING STRATEGY

 |    : /

JANUARY-DECEMBER 2019   -  Volume: 6 -  Pages: [17 p.]

DOI:

https://doi.org/10.6036/NT9175

Authors:

RAUL EDUARDO HUAROTE ZEGARRA
- INDIRA ELIZABETH ROJAS ESCOBAR - VEGA LUJAN

Disciplines:

  • Computer Sciences (ARTIFICIAL INTELLIGENCE / INTELIGENCIA ARTIFICIAL )

Downloads:   123

How to cite this paper:  
Download pdf

Download pdf

Received Date :   21 March 2019

Reviewing Date :   22 March 2019

Accepted Date :   26 April 2019


Key words:
Neocognitrón, Filtro Gabor, bordes orientados, reconocimiento digital de manuscrito, taza de reconocimiento, Neocognitrón, Gabor filter, oriented edges, handwitten digit recognition, recognition cup.
Article type:
ARTICULO DE INVESTIGACION / RESEARCH ARTICLE
Section:
RESEARCH ARTICLES

ABSTRACT
This research modifies the training strategy in the first level of the neocognitron of 2003, using symmetric gabor filters, highlighting the excitatory and inhibitory zones that will allow extracting the characteristics more accurately as weights of the S cells of the first level, improving its recognition rate by 0.5%. To verify the results have been applied to the problem of recognition of handwritten digits, with a set of training of 5000 patterns and with tests of 8000 patterns extracted from the database of digits MNIST (Modified National Institute of Standars and Technology). The results obtained in the experiments show that the maximum recognition rate of the neocognitron with changes in the first level is 99.02%, while the maximum recognition rate for the neocognitron of 2003 is 98.5%. Also, the modified model has an efficiency of 83% with respect to the original model and in 35% of the speed of execution with respect to the original neocognitron, temporal complexity was also reduced by 64%.

Keywords: Neocognitron, Gabor filter, oriented edges, handwitten digit recognition, recognition cup.

Share:  

  • Twittear
  • facebook
  • google+
  • linkedin
  • delicious
  • yahoo
  • myspace
  • meneame
  

Search Content

banner crosscheck

  •  
  • Twitter
  • Twitter
  •  
  • Facebook
  • Facebook
  •  
Tweets por el @revistadyna.
Loading…

Anunciarse en DYNA 

© DYNA New Technologies Journal

EDITORIAL: Publicaciones DYNA SL

Adress: Alameda Mazarredo 69 - 2º, 48009-Bilbao SPAIN

Email: info@dyna-newtech.com - Web: http://www.dyna-newtech.com

 

  • Menu
  • DYNA Publishing
    • DYNA Publishing
    • DYNA
    • DYNA Energy and Sustainability
    • DYNA Management
    • DYNA New Technologies
  • Journal
    • Journal
    • The Journal and its bodies
      • The Journal and its bodies
      • The Journal
      • Editorial Board
      • Advisory-Scientific Board
    • Diffusion & indexation database
    • Mission, Vision & Values
    • Collaborating with DYNA
  • Authors & Referees
    • Authors & Referees
    • Guidelines, rules and forms
    • Collaborating with Journal
  • Papers
    • Papers
    • Search Content
    • Volumes / Issues
    • Most downloaded
    • Sending papers
  • Forum
  • News
    • News
    • News New Technologies
    • Newsletter DNT
  • Advertising
    • Advertising
    • Advertising at DYNA
    • Advertising rates
  • Contact
    • Contact
    • Contacting
  • Search
    • In this Journal
    • Search in DYNA journals
  • Submit
  • Sign in
    • Sign in
    • Privacy Policy

Regístrese en un paso con su email y podrá personalizar sus preferencias mediante su perfil


: *   

: *   

:

: *     

 

  

Loading Loading ...