文件名称:ImprovedPCAFaceRecognitionAlgorithm
-
所属分类:
- 标签属性:
- 上传时间:2012-11-16
-
文件大小:201.03kb
-
已下载:0次
-
提 供 者:
-
相关连接:无下载说明:别用迅雷下载,失败请重下,重下不扣分!
介绍说明--下载内容来自于网络,使用问题请自行百度
摘要:主成分分析(PCA)的人脸识别算法,以减少的特征向量是涉及到对抽象的特点,改进了主成分分析(一)iUumination算法的变化影响酶原sed.The方法是基于上减低与正常化其相应的标准差的特征向量元素相关联的大特征值的特征向量的影响力的想法。耶鲁大学和耶鲁大学面临的数据库面对数据库B是用来验证-Abstract:In principal component analysis(PCA)algorithms for face recognition,to reduce the influence of the
eigenvectors which relate to the changes of the iUumination on abstract features,a modified PCA ( A)
algorithm is propo sed.The method is based on the idea of reducing the influence of the eigenvectors associated
with the large eigenvalues by normalizing the feature vector element by its corresponding standard deviation.
Th e Yale face database and Yale face database B are used to verify the method.The simulation results show
that,f0r front face and even under the condition of limited variation in the facial po ses the proposed method
results in better perform ance than the conventional PCA and linear discriminant analysis(LDA)approaches.and
the computational cost remains the same as that ofthe PCA,and much less than that ofthe LDA.
eigenvectors which relate to the changes of the iUumination on abstract features,a modified PCA ( A)
algorithm is propo sed.The method is based on the idea of reducing the influence of the eigenvectors associated
with the large eigenvalues by normalizing the feature vector element by its corresponding standard deviation.
Th e Yale face database and Yale face database B are used to verify the method.The simulation results show
that,f0r front face and even under the condition of limited variation in the facial po ses the proposed method
results in better perform ance than the conventional PCA and linear discriminant analysis(LDA)approaches.and
the computational cost remains the same as that ofthe PCA,and much less than that ofthe LDA.
(系统自动生成,下载前可以参看下载内容)
下载文件列表
ImprovedPCAFaceRecognitionAlgorithm.PDF
本网站为编程资源及源代码搜集、介绍的搜索网站,版权归原作者所有! 粤ICP备11031372号
1999-2046 搜珍网 All Rights Reserved.