| 一种基于DCT和SVM的人脸识别新方法 |
| |
| 引用本文： | 王孝国,王景玉,黄勇,张雄伟,刘斌.一种基于DCT和SVM的人脸识别新方法[J].军事通信技术,2008(2). |
| |
| 作者姓名： | 王孝国 王景玉 黄勇 张雄伟 刘斌 |
| |
| 作者单位： | 解放军理工大学通信工程学院电子信息工程系;解放军理工大学理学院基础电子学系; |
| |
| 摘 要： | 文中提出了一种新的人脸识别方法,该方法采用DCT提取人脸特征,并采用SVM对该特征进行分类识别。基于该方法,对ORL人脸库进行分类识别,仅用28个特征平均识别率达到97.4%。仿真结果表明,该方法显著地降低了特征维数和计算复杂度,明显提高了特征的可辨别能力,而且,SVM可以有效地提高分类器的分类和推广能力。
|
| 关 键 词： | 人脸识别 离散余弦变换 支持向量机 最近邻分类器 |
Novel Face Recognition Method Based on DCT and SVM |
| |
| WANG Xiao-guo,WANG Jing-yu,HUANG Yong,ZHANG Xiong-wei,LIU Bin.Novel Face Recognition Method Based on DCT and SVM[J].Journal of Military Communications Technology,2008(2). |
| |
| Authors: | WANG Xiao-guo WANG Jing-yu HUANG Yong ZHANG Xiong-wei LIU Bin |
| |
| Institution: | WANG Xiao-guo1,WANG Jing-yu2,HUANG Yong1,ZHANG Xiong-wei1,LIU Bin2(1.Department of Electronic Information Engineering ICE,PLAUST,Nanjing 210007,China,2.Department of Basic Electronics IS,Nanjing 211101,China) |
| |
| Abstract: | A novel face recognition method was presented in this paper.In this method,Discrete Cosine Transform(DCT) was used to extract face features,and Support Vector Machine(SVM) was selected to perform face classification.Experimental results on ORL face database show that the proposed method achieves an average recognition accuracy of 97.4% using only 28 features.The computational complexity and features dimension are reduced greatly,and the dicriminant power of the features is enhanced obviously.Moreover,the cl... |
| |
| Keywords: | face recognition discrete cosine transform support vector machine nearest neighbor classifier |
| 本文献已被 CNKI 等数据库收录！ |
|
