Title :
Distortion invariant selective attention adaptive resonance theory neural network
Author :
Westmacott, Jason ; Lozo, Peter ; Jain, Lakhmi
Author_Institution :
Knowledge-Based Intelligent Eng Syst. Centre, Univ. of South Austalia, Mawson Lakes, SA, Australia
Abstract :
This paper presents an extension to the SAART (selective-attention adaptive resonance theory) neural network to allow distortion invariance. This extended SAART model is now capable of recognising distorted 2D shapes in cluttered environments. This new capability is demonstrated using simulations
Keywords :
ART neural nets; image recognition; invariance; noise; simulation; SAART neural network; cluttered environments; distorted 2D shape recognition; distortion invariance; feedforward excitation-feedback presynaptic facilitation; selective-attention adaptive resonance theory; simulation; Artificial neural networks; Australia; Circuits; Image edge detection; Intelligent systems; Neural networks; Pattern recognition; Resonance; Shape; Subspace constraints;
Conference_Titel :
Knowledge-Based Intelligent Information Engineering Systems, 1999. Third International Conference
Conference_Location :
Adelaide, SA
Print_ISBN :
0-7803-5578-4
DOI :
10.1109/KES.1999.820107
