| M2DPCA与CCLDA相结合的人脸识别 |
| |
| 引用本文： | 冯华丽,刘渊.M2DPCA与CCLDA相结合的人脸识别[J].计算机工程与应用,2014(12):129-132,143. |
| |
| 作者姓名： | 冯华丽 刘渊 |
| |
| 作者单位： | [1]无锡商业职业技术学院教育信息化中心,江苏无锡214153 [2]江南大学数字媒体学院,江苏无锡214122 |
| |
| 基金项目： | 国家自然科学基金（No.60975027）. |
| |
| 摘 要： | CCLDA算法将图像矩阵转化为向量进行处理，该算法易造成数据维数很大，计算量复杂并容易出现“小样本”等问题。针对以上这些问题，提出了一种基于模块化2DPCA和CCLDA相结合的协同处理方法并应用于人脸识别领域。并且在ORL和XM2VTS人脸库上的实验结果表明，新方法在识别效果上有比以往的算法更为明显的优势。
|
| 关 键 词： | 上下文约束 模块化二维主成分分析（M2DPCA） 基于上下文约束线性判别分析（CCLDA） 人脸识别 |
Face recognition algorithm based on modular 2DPCA and CCLDA |
| |
| FENG Huali,LIU Yuan.Face recognition algorithm based on modular 2DPCA and CCLDA[J].Computer Engineering and Applications,2014(12):129-132,143. |
| |
| Authors: | FENG Huali LIU Yuan |
| |
| Affiliation: | 1.Education Information Center, Wuxi Institute of Commerce and Technology, Wuxi, Jiangsu 214153, China 2.School of Digital Media, Jiangnan University, Wuxi, Jiangsu 214122, China) |
| |
| Abstract: | An improved face recognition algorithm is proposed based on the combination of modular 2DPCA and Contex-tual Constraints based Linear Discriminant Analysis（CCLDA）because of the disadvantages of CCLDA. CCLDA first transforms an image matrix to a vector which causes high dimensionality and computational complexity and not considers the local feature. Experimental results obtained on ORL and XM2VTS databases show the effectiveness of the new method. |
| |
| Keywords: | contextual constraints Modular 2-Dimensional Principal Component Analysis（M2DPCA） Contextual Con-straints based Linear Discriminant Analysis（CCLDA） face recognition |
| 本文献已被 CNKI 维普 等数据库收录！ |
