| 融合自适应加权和局部奇异值分解的人脸识别 |
| |
| 引用本文： | 刘敏,李晓东,王振海. 融合自适应加权和局部奇异值分解的人脸识别[J]. 计算机工程与应用, 2009, 45(14): 173-175. DOI: 10.3778/j.issn.1002-8331.2009.14.053 |
| |
| 作者姓名： | 刘敏 李晓东 王振海 |
| |
| 作者单位： | 临沂师范学,院信息学院,山东,临沂,276000;东南大学,自动化学院,复杂工程系统测量与控制教育部重点实验室,南京,210096 |
| |
| 摘 要： | 提出了融合自适应加权和局部奇异值分解的人脸识别方法。首先,对每个训练样本分割出人脸图像的5个特殊区域并分别进行奇异值分解,提取一些较大的奇异值构成每一区域的特征向量。然后,计算各局部块的类内距离平均值和类间距离平均值,从而得到各部分对应的权值。识别阶段,计算待识别人脸图像每一区域对所有训练样本人脸图像相应区域的隶属度,最后采用加权融合策略做出判断。基于ORL和FERET人脸数据库的实验结果表明提出的方法具有有效性和可行性。
|
| 关 键 词： | 人脸识别 奇异值分解 自适应加权 |
| 收稿时间： | 2008-12-12 |
| 修稿时间： | 2009-3-2 |
Face recognition based on adaptive weight and local singular value decomposition |
| |
| LIU Min,LI Xiao-dong,WANG Zhen-hai. Face recognition based on adaptive weight and local singular value decomposition[J]. Computer Engineering and Applications, 2009, 45(14): 173-175. DOI: 10.3778/j.issn.1002-8331.2009.14.053 |
| |
| Authors: | LIU Min LI Xiao-dong WANG Zhen-hai |
| |
| Affiliation: | 1.School of Information,Linyi Normal University,Linyi,Shandong 276000,China 2.Key Lab. of Measurement and Control of Complex Sys. of Eng.,School of Automation,Southeast University,Nanjing 210096,China |
| |
| Abstract: | A face recognition method based on adaptive weight and local singular value decomposition is proposed in this paper. Firstly,each facial image in the training sample set is divided into five special regions,and singular value decomposition is performed on these regions.Some bigger singular values are taken to form the feature vector of each region,subsequently.Secondly, the same approach as what has given above is used to get the feature vectors of all regions in the testing facial image.Weight value that b... |
| |
| Keywords: | face recognition Singular Value Decomposition（SVD） adaptive weight |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载免费的PDF全文 |
|
