Design of binary serial-coded-filters by using Hopfield model

被引：0

作者：

Liu, Ying ^{[1
]}

Lu, Mingzhe ^{[1
]}

Zhang, Jianming ^{[1
]}

Fang, Zhiliang ^{[1
]}

Liu, Fulai ^{[1
]}

Mu, Guoguang ^{[1
]}

机构：

[1] Nankai Univ, Tianjin, China

来源：

Hongwai Yu Haomibo Xuebao/Journal of Infrared and Millimeter Waves | 1994年 / 13卷 / 03期

关键词：

Artificial intelligence - Computer simulation - Neural networks;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Binary serial-coded-filters were designed based on the optimization function of the Hopfield model. Compared with other methods, this method needs less filters to carry out the same recognition task. A computer simulation has been conducted with this method. The 26 English letters were used as the target objects. The simulation results were presented and discussed. It was demonstrated that this method has very strong error-tolerant capacity, all target objects can be correctly recognized when the characteristic codes are properly chosen. The characteristic codes have some influence on the storage and error-tolerant capacity of the filter.

引用

页码：221 / 226