| 一种基于2DPCA和DLDA的人脸识别方法 |
| |
| 引用本文： | 王杨峰,付永庆.一种基于2DPCA和DLDA的人脸识别方法[J].应用科技,2007,34(7):31-33. |
| |
| 作者姓名： | 王杨峰 付永庆 |
| |
| 作者单位： | 哈尔滨工程大学,信息与通信工程学院,黑龙江,哈尔滨,150001 |
| |
| 摘 要： | 一维方法特征提取时运算量大，图像较大时很不方便．二维的方法特征提取直接，速度快，但提取出的特征是矩阵，特征数量大，影响分类速度．结合2者的优点，提出二维与一维相结合的特征提取方法来识别人脸．先用二维PCA（2DPCA）处理原始图像，降维后进行DLDA处理．在ORL人脸库中验证了这种算法的可行性，结果表明识别率和分类速度均有提高．
|
| 关 键 词： | 人脸识别 特征提取 2维主成分分析 直接线性判刷分析 |
| 文章编号： | 1009-671X（2007）07-0031-03 |
| 修稿时间： | 2007-01-15 |
A face recognition method based on two-dimensional PCA and DLDA |
| |
| WANG Yang-feng,FU Yong-qing.A face recognition method based on two-dimensional PCA and DLDA[J].Applied Science and Technology,2007,34(7):31-33. |
| |
| Authors: | WANG Yang-feng FU Yong-qing |
| |
| Institution: | College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China |
| |
| Abstract: | Computational amount is large when using the one-dimensional method of feature extraction, especially for large images. The two-dimensional method extracts feature directly and rapidly, but the features must be expressed with matrixes. Therefore the classification speed is affected by too many features. Hence, a new approach is proposed to recognize the human faces in combination with one dimensional and two dimensional feature extraction advantages. First, 2DPCA is used to deal with original images, then DLDA is used to compress the feature. The feasibility is verified in ORL face database. The results show that both recognition rate and classification speed are improved. |
| |
| Keywords: | face recognition feature extraction 2DPCA DLDA |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
