摘要
文章针对大样本集的手写体汉字数据,指出了改进SDM 模型的必要性,提出了两点改进措施,并在此基础上采用面向对象的程序设计方法,构造了一个手写体汉字联想记忆系统。实验表明:该系统具有较好的联想记忆性能。
Asfarasthelargesamplesetof handwritten Chinese characters, it is necessary to improve the performance of Kanerva’s SDM model. Two modified methods are proposed in this paper. Based on the improved SDM model, a handwritten Chinese character associative memory system is designed with Object - oriented programming. Experiments have shown thatthe performance ofthis associative memory system is good.
出处
《微电子学与计算机》
CSCD
北大核心
2000年第1期1-4,共4页
Microelectronics & Computer
基金
国家自然科学基金资助
关键词
联想记忆
手写体汉字识别
SDM模型
文字识别
Sparse distributed memory ( SDM) , Associative memory ,Handwritten Chinese character recognition