| 大间距无监督正交特征提取算法 |
| |
| 引用本文： | 林玉娥,李敬兆,梁兴柱,林玉荣.大间距无监督正交特征提取算法[J].传感器与微系统,2012,31(4):143-145. |
| |
| 作者姓名： | 林玉娥 李敬兆 梁兴柱 林玉荣 |
| |
| 作者单位： | 1. 安徽理工大学计算机科学与工程学院,安徽淮南,232001
2. 哈尔滨工业大学航天学院,黑龙江哈尔滨,150001 |
| |
| 基金项目： | 国家自然科学基金资助项目(60975009,61170060);安徽省自然科学基金资助项目(1208085QF123) |
| |
| 摘 要： | 以主成分分析和局部保持投影为理论基础,提出了一种同时考虑数据样本的全局和局部特性的大间距无监督正交特征提取算法,算法的目标函数采用大间距准则,避免了由于矩阵求逆带来的小样本问题,同时为了进一步增强算法的识别性能,对所求取的投影矩阵进行了正交化约束,最后人脸库上的实验结果表明所提方法的有效性.
|
| 关 键 词： | 大间距 特征提取算法 目标函数 小样本问题 |
Maximum margin unsupervised orthogonal feature extraction algorithm |
| |
| LIN Yu-e
,
LI Jing-zhao
,
LIANG Xing-zhu
,
LIN Yu-rong.Maximum margin unsupervised orthogonal feature extraction algorithm[J].Transducer and Microsystem Technology,2012,31(4):143-145. |
| |
| Authors: | LIN Yu-e LI Jing-zhao LIANG Xing-zhu LIN Yu-rong |
| |
| Affiliation: | 1.School of Computer Science & Engineering,Anhui University of Science and Technology,Huainan 232001,China; 2.School of Astronautics,Harbin Institute of Technology,Harbin 150001,China) |
| |
| Abstract: | A maximum margin unsupervised orthogonal feature extraction algorithm based on principal component analysis and locality preserving projection is proposed,in which both global and local features of the data samples are taken into account.The proposed method adopts the maximum margin criterion as object function and avoids the small sample size problem.To further enhance the recognition performance of the algorithm,orthogonal constraint projection matrix is given.Experimental results on face database demonstrate the effectiveness of the proposed method. |
| |
| Keywords: | maximum margin feature extraction algorithm objective function small sample size problem |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
