| 模糊标号典型相关分析及其在人脸识别中应用 |
| |
| 引用本文： | 苏志勋,刘艳艳,刘秀平.模糊标号典型相关分析及其在人脸识别中应用[J].大连理工大学学报,2009,49(1):133-138. |
| |
| 作者姓名： | 苏志勋 刘艳艳 刘秀平 |
| |
| 作者单位： | 1. 大连理工大学应用数学系,辽宁大连,116024
2. 防灾科技学院基础部,河北,三河,065201 |
| |
| 基金项目： | 新世纪优秀人才支持计划资助项目(NCET-05-0275);;国家自然科学基金资助项目(60673006,60873181) |
| |
| 摘 要： | 把样本分布信息融于特征提取过程将有助于提高特征的分类能力.利用模糊隶属度概念,提出一种基于模糊标号典型相关分析的特征提取新方法.构造模糊标号刻画样本的分布情况,并将其与典型相关分析结合,能提取综合灰度信息和分布信息的有效判别特征.此外,针对样本不足导致的小特征值包含较多干扰信息的问题,基于矩阵理论及双空间分析思想,进一步提出双空间模糊标号典型相关分析算法,缓解了过小特征值对算法性能的影响.在ORL和组合人脸数据库上的实验结果表明新特征具有良好的分类能力,证实了所提算法的有效性及应用价值.
|
| 关 键 词： | 典型相关分析 模糊隶属度 小样本问题 特征提取 人脸识别 |
Fuzzy label canonical correlation analysis and its application to face recognition |
| |
| SU Zhixun,LIU Yanyan,LIU Xiuping.Fuzzy label canonical correlation analysis and its application to face recognition[J].Journal of Dalian University of Technology,2009,49(1):133-138. |
| |
| Authors: | SU Zhixun LIU Yanyan LIU Xiuping |
| |
| Institution: | SU Zhi-xun1,LIU Yan-yan2,LIU Xiu-ping11.Department of Applied Mathematics,Dalian University of Technology,Dalian 116024,China,2.Department of Basic Courses,Institute of Disaster Prevention Science , Technology,Sanhe 065201 |
| |
| Abstract: | Incorporating the sample distribution information into the process of feature extraction is beneficial to promoting the classification performance of features.A fuzzy label canonical correlation analysis(CCA) algorithm is proposed for image feature extraction.Fuzzy class labels in the form of membership degrees are designed elaborately to represent the sample distribution.Then the fuzzy labels are embedded in CCA to extract more discriminative features which combine the information about gray level and dist... |
| |
| Keywords: | canonical correlation analysis fuzzy membership degree small sample size problem feature extraction face recognition |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《大连理工大学学报》浏览原始摘要信息 |
| 点击此处可从《大连理工大学学报》下载免费的PDF全文 |
|
