Title :
Modular mART for 3D target recognition
Author :
Kim, Eun-Soo ; Cha, Jin-Woo ; Ryu, Chung-Sang
Author_Institution :
Dept. of Electron. Eng., Kwangwoon Univ., Seoul, South Korea
fDate :
7/31/1997 12:00:00 AM
Abstract :
A modified adaptive resonance theory (mART) neural network of modular structure is proposed. The similarity function and weight resolution of the ART neural networks are modified, and the cluster merging algorithm and modular training method are both introduced. The results from 3D target recognition experiments are compared with those of a self-organising map (SOM) and single mART
Keywords :
ART neural nets; learning (artificial intelligence); object recognition; target tracking; 3D target recognition; cluster merging algorithm; mART; modified adaptive resonance theory; modular training method; neural network; similarity function; weight resolution;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19970908
