Title :
An invariant pattern recognition machine using a modified ART architecture
Author :
Srinivasa, Narayan ; Jouaneh, Musa
Author_Institution :
Dept. of Mech. Eng., Florida Univ., Gainesville, FL, USA
Abstract :
A novel invariant pattern recognition machine is proposed based on a modified ART architecture. Invariance is achieved by adding a new layer called F3, beyond the F2 layer in the ART architecture. The design of the weight connections between the nodes of the F2 layer and the cells of the F3 layer are similar to the invariance net. Computer simulations show that the model is not only invariant to translations and rotations of 2-D binary images but also noise-tolerant to these transformed images
Keywords :
image recognition; invariance; neural nets; parallel architectures; pattern recognition equipment; 2D binary images; ART architecture; adaptive resonance theory; invariance net; invariant pattern recognition machine; neural nets; nodes; noise tolerance; weight connections; Brain modeling; Computer architecture; Computer simulation; Filters; Holography; Mechanical engineering; Multilayer perceptrons; Neural networks; Pattern recognition; Subspace constraints;
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
