摘要
就基于PCA与LDA变换的传统人脸识别方法识别率低但特征提取过程中维数低和基于K-L变换的仿生人脸识别方法识别率高但在特征提取过程中维数过高的的问题,将两者的优点相结合,提出了一种基于PCA与LDA变换的仿生人脸识别新方法。通过PCA与LDA变换对训练人脸样本进行特征提取,然后构建各类样本的覆盖区域。再通过判断待识别人脸特征在各覆盖区域的归属情况来识别人脸。实验收到了预期的效果,证明了方法的可行性。
A new method of biomimetic pattern face recognition theory based on PCA and LDA transform is proposed.This method has solved the low recognition rate and the excessively high dimension problem,the features of human face on the training samples are extracted through PCA and LDA,and are used to construct the cover region of each kind of sample.The person face is distinguished through the judgment that the person face characteristic belongs to which kind of cover region or doesn＇t not belong to any region.The experiment has received the anticipated effect,and has proven this method feasibility.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第19期160-163,共4页
Computer Engineering and Applications
基金
湖南省自然科学基金(No.06jj50109)
湖南省科技计划项目(No.06fj3161)~~
关键词
主成分分析(PCA)
线性鉴别分析(LDA)
K-L变换
仿生模式识别
高维空间几何形体
同源连续性
Principal Component Analysis（PCA）
Linear Discriminant Analysis（LDA）
K-L transformation
biomimetic pattern recognition
high dimentional space geometric solid
Homologous continuity
