| 稀疏判决分析在表情识别中的应用 |
| |
| 引用本文： | 黄勇.稀疏判决分析在表情识别中的应用[J].计算机工程与应用,2012,48(7):172-173,211. |
| |
| 作者姓名： | 黄勇 |
| |
| 作者单位： | 柳州铁道职业技术学院电子工程系,广西柳州,545007 |
| |
| 摘 要： | 提出了一种基于稀疏判决分析的人脸表情识别方法,称之为SDA。SDA引入稀疏表述并结合半监督判决分析SSDA,通过稀疏重构处理,获得图像的局部结构信息,且由于稀疏表述本身具有的判决性,SDA只需很少样本就能获得较好的效果。基于JAFFE和CED-WYU两个表情数据库的识别结果表明,基于SDA的特征提取方法能有效地提高识别率。
|
| 关 键 词： | 稀疏表述 线性判决分析 半监督判决分析 稀疏判决分析 表情识别 |
Sparse discriminant analysis for expression recognition |
| |
| HUANG Yong.Sparse discriminant analysis for expression recognition[J].Computer Engineering and Applications,2012,48(7):172-173,211. |
| |
| Authors: | HUANG Yong |
| |
| Affiliation: | HUANG Yong Department of Electronic Engineering, Liuzhou Railway Vocational Technical College, Liuzhou, Guangxi 545007, China |
| |
| Abstract: | A facial expression recognition method based on Sparse Discriminant Analysis（SDA）is proposed in this paper. The graph in SDA is constructed by sparse representation and incorporate Semi-Supervised Discriminant Analysis（SSDA）, thus the local structure information is automatically modeled, and with the natural discriminative power of sparse representation, SDA can get better performance only resorting to few extra unlabeled samples. Experimental result on JAFFE and CED-WYU show that SDA is an effective method for improving the recognition accuracy. |
| |
| Keywords: | sparse representation Linear Discriminant Analysis（LDA） Semi-Supervised Discriminant Analysis（SSDA） Sparse Discriminant Analysis（SDA） expression recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
