| 基于双向PCA和K近邻的人脸识别算法 |
| |
| 引用本文： | 王心醉,李岩,郭立红,肖永鹏,董宁宁,杨丽梅.基于双向PCA和K近邻的人脸识别算法[J].解放军理工大学学报,2010(6):623-627. |
| |
| 作者姓名： | 王心醉 李岩 郭立红 肖永鹏 董宁宁 杨丽梅 |
| |
| 作者单位： | 中国科学院长春光学精密机械与物理研究所;中国科学院研究生院;中国科学院长春光学精密机械与物理研究所;中国科学院长春光学精密机械与物理研究所;东北师范大学计算机学院;中国科学院长春光学精密机械与物理研究所;长春工业大学机电工程学院 |
| |
| 基金项目： | 中国科学院知识创新工程领域前沿资助项目;长春市科技攻关项目(07163UC070) |
| |
| 摘 要： | 针对当前人脸识别算法的运行速度和识别准确率的矛盾,提出一种基于双向主成分法(Bidirectional PCA,BD-PCA)和K近邻法K-NN(K-nearest neighbor)的人脸识别算法。在VC6.0平台下基于ORL人脸库进行实验,首先利用双向PCA算法对训练人脸样本和测试人脸样本进行方向和列方向降维并提取特征脸;然后用K近邻方法对特征脸进行人脸匹配。实验表明,提出的算法在有效降低运算时间的同时,又能取得很高的识别准确率,具有一定的可行性。
|
| 关 键 词： | 双向PCA K-NN 人脸识别 |
Face recognition algorithm based on BD-PCA and K-NN |
| |
| WANG Xin-zui,LI Yan,GUO Li-hong,XIAO Yong-peng,DONG Ning-ning and YANG Li-mei.Face recognition algorithm based on BD-PCA and K-NN[J].Journal of PLA University of Science and Technology(Natural Science Edition),2010(6):623-627. |
| |
| Authors: | WANG Xin-zui LI Yan GUO Li-hong XIAO Yong-peng DONG Ning-ning and YANG Li-mei |
| |
| Institution: | Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;Graduate School of Chinese Academy of Sciences,Beijing 100039,China;Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;School of Computer Science,Northeast Normal University,Changchun 130117,China;Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;School of Electro-mechanic Engineering,Changchun University of Technology,Changchun 130033,China |
| |
| Abstract: | Aiming at the cont radict ion betw een the running t ime and the accuracy of the existing face
recog nit ion algo rithm, a face recog nit ion algo rithm based on BD-PCA and K-NN w as proposed. T he
algo rithm w as tested on the ORL face database at VC 6. 0. Fir st ly, the dimension o f the row and the
co lumn of t raining and testing samples w as reduced and the Eigenface w as o btained using the BD-PCA
algo rithm. Then, the Eig enface w as matched using the K-NN alg orithm. The exper imental result
indicates that the alg orithm can g et high accuracy while greatly reducing the running t ime, w hich pro ves
the feasibility of the algo rithm |
| |
| Keywords: | BD-PCA K-NN face reco gnit ion |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《解放军理工大学学报》浏览原始摘要信息 |
| 点击此处可从《解放军理工大学学报》下载免费的PDF全文 |
|
