摘要
研究基于不同模式特征的多分类器联合问题 ,提出一种新的分类器联合方法。将该方法应用于联机几何图形的识别 ,实验中联合了 3种分类器 ,每种分类器的机制各不相同 ,并且都基于不同的模式特征。将该分类器联合方法与现有的几种联合方法进行实验比较 ,实验结果表明该方法具有较高的识别率。
The problem of combining multiple classifiers, which are different in mechanism and based on different pattern features is studied. A new scheme for classifier combination is proposed. The scheme is applied to an on-line graphics recognition problem in which there are three classifiers to be combined. The scheme is compared with several other existing combination schemes experimentally. The results show that the scheme has a higher classification rate than others.
出处
《控制与决策》
EI
CSCD
北大核心
2003年第1期92-95,共4页
Control and Decision
基金
国防预研基金资助项目 (6 .6 .6 .5 )
陕西省自然科学基金资助项目 (99X14)
