| 基于非线性特征提取和SVM的人脸识别算法 |
| |
| 引用本文： | 孙大瑞,吴乐南.基于非线性特征提取和SVM的人脸识别算法[J].电子与信息学报,2004,26(2):307-311. |
| |
| 作者姓名： | 孙大瑞 吴乐南 |
| |
| 作者单位： | 东南大学无线电工程系,南京,210096;东南大学无线电工程系,南京,210096 |
| |
| 摘 要： | 传统的PCA或LDA都是从像素的二阶依赖上考虑的，对于多像素之间的依赖性或像素的高阶关系不敏感。该文利用核函数方法提取像素高阶相关，并与线性SVM相结合来进行人脸识别。从Yale人脸库上的实验结果可以看出，非线性特征提取是很有效的，并且SVM分类器的性能优于最近邻分类器。
|
| 关 键 词： | 非线性特征提取 KPCA KDA 线性SVM |
| 文章编号： | 1009-5896(2004)02-0307-05 |
| 收稿时间： | 2002-10-7 |
| 修稿时间： | 2002年10月7日 |
Face Recognition Based on Nonlinear Feature Extraction and SVM |
| |
| Sun Da-rui,Wu Le-nan.Face Recognition Based on Nonlinear Feature Extraction and SVM[J].Journal of Electronics & Information Technology,2004,26(2):307-311. |
| |
| Authors: | Sun Da-rui Wu Le-nan |
| |
| Affiliation: | Department of Radio Engineering Southeast University Nanjing 210096 China |
| |
| Abstract: | Both PCA and LDA are performed by only using the second-order statistics among image pixels, and not sensitive to high order statistics in the data. In this paper, the kernel function method is used to extract the high order relations, and the Linear Support Vector Machines (LSVM) is selected to perform the face classification. The experiment on Yale face database shows that the nonlinear feature extraction method is effective, and SVM is better than nearest neighbor classifier. |
| |
| Keywords: | Nonlinear feature extraction KPCA KDA Linear SVM |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《电子与信息学报》浏览原始摘要信息 |
|
点击此处可从《电子与信息学报》下载全文 |
