Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Invariant Pattern Recognition of 2D Images Using Neural Networks and FrequencyDomain Representation \Lambda
 

Summary: Invariant Pattern Recognition of 2D Images Using Neural Networks
and Frequency­Domain Representation \Lambda
Fernando C'esar C. De Castro
decastro@ee.pucrs.br
Jos'e Nelson Amaral
amaral@ee.pucrs.br
Paulo Roberto G. Franco
pfranco@ee.pucrs.br
Electrical Engineering Department
Pontif'icia Universidade Cat'olica do Rio Grande do Sul
90619­900 ­ Porto Alegre ­ RS ­ Brazil
Abstract
Frequency domain representation of two dimensional gray­level images is used to develop a pattern
recognition method that is invariant to rotation, translation and scaling. Frequency domain representa­
tion is a natural feature detector that allows the use of only few directions of highest energy as training
data for a set of Artificial Neural Networks (ANNs). We developed a new algorithm that uses the spec­
tral information stored in these ANNs to compare a given image with a known pattern, determining the
relative translation between them and yielding a measure of their similarity. The representation and
method we adopted has the advantage of leaving only the rotation of the object as a free parameter to
be determined by the algorithm. We minimize the spectral resolution noise using Spectral Directional

  

Source: Amaral, José Nelson - Department of Computing Science, University of Alberta

 

Collections: Computer Technologies and Information Sciences