| 基于融合两类核Fisher鉴别函数的人脸识别方法 |
| |
| 引用本文： | 成新民,蒋云良.基于融合两类核Fisher鉴别函数的人脸识别方法[J].武汉理工大学学报(交通科学与工程版),2009,33(2). |
| |
| 作者姓名： | 成新民 蒋云良 |
| |
| 作者单位： | 1. 湖州师范学院信息工程学院,湖州,313000
2. 湖州师范学院信息工程学院,湖州,313000;浙江大学计算机科学与技术学院,杭州,310027 |
| |
| 基金项目： | 国家自然科学基金，浙江省自然科学基金 |
| |
| 摘 要： | 提出了一种基于两类核Fisher鉴别分析(KFDA)的人脸识别方法,对每2个不同人脸类别求解一个核Fisher鉴别函数,其优点是能针对特定的2个人脸图像类别,抽取区分该2类人脸的最佳鉴别特征,克服了多类KFDA和2类KFDA相比是次优的问题.为解决KFDA计算量大的问题,将MSE推广为基于核的MSE(KMSE),用其得到核Fisher鉴别函数,减少了训练和识别的计算时间.在识别阶段应用了两种融合方法融合各个基于KMSE的核Fisher鉴别函数.
|
| 关 键 词： | 人脸识别 两类核Fisher鉴别分析 最小平方误差方法 核方法 判别融合 |
Face Recognition Based on Combining Two-class Kernel Fisher Discriminant Functions |
| |
| Cheng Xinmin,Jiang Yunliang.Face Recognition Based on Combining Two-class Kernel Fisher Discriminant Functions[J].journal of wuhan university of technology(transportation science&engineering),2009,33(2). |
| |
| Authors: | Cheng Xinmin Jiang Yunliang |
| |
| Institution: | School of Information Engineering;Huzhou Teachers College;Huzhou 313000;College of Computer Science;Zhejiang University;Hangzhou 310027 |
| |
| Abstract: | Multiclass Fisher discriminant analysis(MFDA) can extract Fisher optimal discriminant features.However MFDA is suboptimal compared with two-class FDA.To overcome the problem of MFDA,a face recognition method based on two-class FDA is proposed in this paper.This method can extract nonlinear Fisher optimal discriminant features distinguishing two specific classes of face images.Due to FLDA equivalent to minimum squared error(MSE) method,kernel based MSE is proposed to obtain kernel Fisher discriminant functio... |
| |
| Keywords: | face recognition two-class kernel Fisher discriminant analysis minimum squared error method kernel method decision level fusion |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
