Title :
A genetic algorithm for training image classification neural networks
Author :
Zhang, Ching ; Wang, Fangju
Author_Institution :
Dept. of Syst. Design Eng., Waterloo Univ., Ont., Canada
Abstract :
Neural networks are becoming effective tools for digital image classification. They have advantages including simple and flexible structures and higher tolerance to errors. The major drawbacks which limit neural networks for practical applications include slow training phase and divergence of training. In this research, a new method has been developed to address the drawbacks. This method is based on genetic algorithms
Keywords :
genetic algorithms; image classification; learning (artificial intelligence); neural nets; digital image; genetic algorithm; image classification; learning; neural networks; training phase; Computer networks; Design engineering; Digital images; Flexible structures; Genetic algorithms; Image classification; Information science; Neural networks; Pattern classification; Systems engineering and theory;
Conference_Titel :
Systems, Man, and Cybernetics, 1994. Humans, Information and Technology., 1994 IEEE International Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-2129-4
DOI :
10.1109/ICSMC.1994.400198
