Welcome!
To use the personalized features of this site, please log in or register.
If you have forgotten your username or password, we can help.
My Menu
Saved Items

Pattern Recognition

Rotation-Invariant Pattern Recognition: A Procedure Slightly Inspired on Olfactory System and Based on Kohonen Network

M. B. Palermo1 and L. H. A. Monteiro1, 2 Contact Information

(1)  Universidade Presbiteriana Mackenzie, Pós-graduação em Engenharia Elétrica, Escola de Engenharia, Rua da Consolação, n.896, 01302-907, São Paulo, SP, Brazil
(2)  Universidade de São Paulo, Departamento de Engenharia de Telecomunicações e Controle, Escola Politécnica, Av Prof. Luciano Gualberto, travessa 3, n.380, 05508-900, São Paulo, SP, Brazil
Abstract
A computational scheme for rotation-invariant pattern recognition based on Kohonen neural network is developed. This scheme is slightly inspired on the vertebrate olfactory system, and its goal is to recognize spatiotemporal patterns produced in a two-dimensional cellular automaton that would represent the olfactory bulb activity when submitted to odor stimuli. The recognition occurs through a multi-layer Kohonen network that would represent the olfactory cortex. The recognition is invariant to rotations of the patterns, even when a noise lower than 1% is added.

Contact Information L. H. A. Monteiro
Email: luizm@mackenzie.br
Fulltext Preview (Small, Large)
Image of the first page of the fulltext


Export this chapter
Export this chapter as RIS | Text
 
Remote Address: 38.107.191.113 • Server: mpweb06
HTTP User Agent: CCBot/1.0 (+http://www.commoncrawl.org/bot.html)