摘要
本文研究了人脸识别应用中的"维数灾难"问题。针对经典欧式距离不能较好刻画高维空间中样本间相似性特征,我们引入了相关系数鉴别分析方法,在高维空间中,基于相关系数提取鉴别特征,实现了人脸样本的维数约简。本文实验结果表明,该方法在小规模及大规模数据集上均具有较好的性能。
This paper considers the dimensionality problem in face recognition application.The classic Euclidean distance is so simple that it can not describe the similarity relationships between samples in high-dimension space.The Correlation Discriminant Analysis method is introduced.In high-dimension space,the correlation similarity features are extracted.Finally,Random Forest is adopted as the classifier.The experiments in the paper valuate the performance of the proposed method on large databases and small databases.
出处
《安徽电子信息职业技术学院学报》
2012年第1期15-18,共4页
Journal of Anhui Vocational College of Electronics & Information Technology
基金
琼台师范高等专科学校科研项目(批准号:qtky2010-20)
关键词
相关系数鉴别分析
随机森林
人脸识别
correlation discriminant analysis random forest face recognition
