| 基于二维主分量分析的面部表情识别 |
| |
| 引用本文： | 程剑,应自炉.基于二维主分量分析的面部表情识别[J].计算机工程与应用,2006,42(5):32-33,39. |
| |
| 作者姓名： | 程剑 应自炉 |
| |
| 作者单位： | 五邑大学信息学院,广东,江门,529020 |
| |
| 摘 要： | 提出了一种直接基于图像矩阵的二维主分量分析(2DPCA)和多分类器联合的面部表情识别方法。首先利用2DPCA进行特征提取,然后用基于模糊积分的多分类器联合的方法对七种表情(生气、厌恶、恐惧、高兴、中性、悲伤、惊讶)进行识别。在JAFFE人脸表情静态图像库上进行实验,与传统主分量分析(PCA)相比,采用2DPCA进行特征提取,不仅识别率比较高,而且运算速度也有很大的提高。
|
| 关 键 词： | 二维主分量分析 主分量分析 人脸表情识别 特征提取 |
| 文章编号： | 1002-8331-（2006）05-0032-02 |
| 收稿时间： | 2005-07 |
| 修稿时间： | 2005-07 |
Facial Expression Recognition Based on 2DPCA |
| |
| Cheng Jian,Ying Zilu.Facial Expression Recognition Based on 2DPCA[J].Computer Engineering and Applications,2006,42(5):32-33,39. |
| |
| Authors: | Cheng Jian Ying Zilu |
| |
| Affiliation: | Information School,Wuyi University,Jiangmen,Guangdong 529020 |
| |
| Abstract: | A method of two-dimensional principal component analysis(2DPCA) which is based on image matrices to directly construct the image total scatter matrix and multiclassifier combination is proposed for facial expression recognition.First,2DPCA is applied to extract features of images,then seven expressions(angry,disgust,fear,happy,neutral,sad and surprise) recognition is implemented based on multiclassifier combination with fuzzy integral.The experimental results on JAFFE facial expression database show that 2DPCA used for feature extraction is more powerful than classical PCA. |
| |
| Keywords: | 2DPCA PCA facial expression recognition feature extraction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
