| 一种核主元分析的人脸识别方法及其DSP实现 |
| |
| 引用本文： | 刘 嵩,谭建军. 一种核主元分析的人脸识别方法及其DSP实现[J]. 电讯技术, 2012, 52(8): 1320-1323 |
| |
| 作者姓名： | 刘 嵩 谭建军 |
| |
| 作者单位： | 湖北民族学院信息工程学院,湖北恩施,445000 |
| |
| 摘 要： | 针对核主元分析在人脸识别中的时间开销过大的不足,提出了一种核主元分析的DSP实现方法.介绍了核主元分析算法原理,给出了DSP平台的硬件框架和程序优化方案.在DSP硬件平台上的测试结果表明:系统运行稳定,与通用计算机平台相比,明显提高了识别速度,同时保证了人脸识别率,具有一定的实用价值.
|
| 关 键 词： | 人脸识别 特征提取 核主元分析 优化程序 数字信号处理 |
A face recognition method based on KPCA and its DSP implementation |
| |
| LIU Song and TAN Jian-jun. A face recognition method based on KPCA and its DSP implementation[J]. Telecommunication Engineering, 2012, 52(8): 1320-1323 |
| |
| Authors: | LIU Song and TAN Jian-jun |
| |
| Affiliation: | College of Information Engineering,Hubei University for Nationalities,Enshi 445000,China;College of Information Engineering,Hubei University for Nationalities,Enshi 445000,China |
| |
| Abstract: | In order to reduce recognition time of kernel principle component analysis(KPCA) algorithm,the realization scheme of KPCA based on DSP is presented in this paper.The KPCA algorithm is introduced,hardware platform frame based on DSP and program optimization scheme are given.The experimental results in DSP platform show that the system is of good stability,high recognition rate and high speed,practical value,and it shows obvious advantage over that of universal computer. |
| |
| Keywords: | face recognition feature extract kernel principle component analysis optimization program DSP |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《电讯技术》浏览原始摘要信息 |
|
点击此处可从《电讯技术》下载全文 |
|
