| 注册
首页|期刊导航|半导体学报|一种可编程Hamming神经网络特征提取器

一种可编程Hamming神经网络特征提取器

栗国星 石秉学

半导体学报2001,Vol.22Issue(3):378-382,5.
半导体学报2001,Vol.22Issue(3):378-382,5.

一种可编程Hamming神经网络特征提取器

A Programmable Hamming Neural Network Feature Extractor

栗国星 1石秉学1

作者信息

  • 1. 清华大学微电子学研究所,
  • 折叠

摘要

Abstract

A current-mode programmable local feature extractor for handwritten digit recognition is put forward based on the principle of Hamming neural network.The templates of the extractor is programmable and their contents can be refreshed according to different requirement.The way of feature merging is also reconfigurable so as to merge different features into different feature classes.A prototype chip of the feature extractor has been implemented in single poly,double metals 1.2μm digital CMOS technology and both the simulated and measured results prove that the extractor can realize the feature extraction well.

关键词

可编程/Hamming神经网络/特征提取/CMOS

分类

信息技术与安全科学

引用本文复制引用

栗国星,石秉学..一种可编程Hamming神经网络特征提取器[J].半导体学报,2001,22(3):378-382,5.

半导体学报

OA北大核心CSCD

1674-4926

访问量2
|
下载量0
段落导航相关论文